Skip to contentSkip to conversation

treewide: Fix a bunch of shift overflows

PATCH Applied Borislav Petkov 13 participants 43 messages Apr 5, 2022 – Jun 24, 2022

Subsystems (21)
lore
Share & tools
lore
#0 lore raw

cover treewide: Fix a bunch of shift overflows

From: Borislav Petkov <[email protected]>

Hi all,

so this is the result of me trying to make allmodconfig actually build here.

Due to some recent changes which added -fsanitize-shift to the build options of an allmodconfig, it started failing here with an old gcc because getting an overflow while shifting is undefined C99 behavior.

gcc warns/errors out with -Werror about this only on newer versions where -pedantic is present while older ones do so even without it. The whole details here:

https://lore.kernel.org/r/YkwQ6%[email protected]

Fixing all those is trivial so please pick up at your convenience.

In order to avoid spamming people unnecessarily, I'm not CCing everyone on each patch but only the relevant maintainers and lists.

Thx.

Borislav Petkov (11):
scsi: aacraid: Fix undefined behavior due to shift overflowing the
constant
ALSA: usb-audio: Fix undefined behavior due to shift overflowing the
constant
bnx2x: Fix undefined behavior due to shift overflowing the constant drm/r128: Fix undefined behavior due to shift overflowing the constant
i2c: ismt: Fix undefined behavior due to shift overflowing the
constant
brcmfmac: sdio: Fix undefined behavior due to shift overflowing the
constant
usb: typec: tcpm: Fix undefined behavior due to shift overflowing the
constant
mt76: Fix undefined behavior due to shift overflowing the constant perf/imx_ddr: Fix undefined behavior due to shift overflowing the
constant
IB/mlx5: Fix undefined behavior due to shift overflowing the constant drm/i915: Fix undefined behavior due to shift overflowing the constant

 .../gpu/drm/i915/gt/uc/abi/guc_actions_abi.h   |  2 +-
 .../i915/gt/uc/abi/guc_communication_ctb_abi.h |  2 +-
 .../gpu/drm/i915/gt/uc/abi/guc_messages_abi.h  |  2 +-
 drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h     |  2 +-
 drivers/gpu/drm/i915/i915_reg.h                | 18 +++++++++---------
 drivers/gpu/drm/r128/r128_drv.h                |  4 ++--
 drivers/i2c/busses/i2c-ismt.c                  |  4 ++--
 .../net/ethernet/broadcom/bnx2x/bnx2x_reg.h    |  2 +-
 .../broadcom/brcm80211/brcmfmac/sdio.c         |  2 +-
 .../net/wireless/mediatek/mt76/mt76x2/pci.c    |  2 +-
 drivers/perf/fsl_imx8_ddr_perf.c               |  2 +-
 drivers/scsi/aacraid/aacraid.h                 |  2 +-
 include/linux/mlx5/port.h                      |  2 +-
 include/linux/usb/pd_bdo.h                     |  2 +-
 sound/usb/usbaudio.h                           |  2 +-
 15 files changed, 25 insertions(+), 25 deletions(-)
Signature · 1 line
2.35.1
#1 lore raw

PATCH 1/11 scsi: aacraid: Fix undefined behavior due to shift overflowing the constant

From: Borislav Petkov <[email protected]>

Fix

drivers/scsi/aacraid/commsup.c: In function ‘aac_handle_sa_aif’:
drivers/scsi/aacraid/commsup.c:1983:2: error: case label does not reduce to an integer constant

    case SA_AIF_BPCFG_CHANGE:
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Adaptec OEM Raid Solutions <[email protected]>
Cc "James E.J. Bottomley" <[email protected]>
Cc "Martin K. Petersen" <[email protected]>
---
 drivers/scsi/aacraid/aacraid.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/aacraid/aacraid.h +1 -1
index f849e7c9d428..5e115e8b2ba4 100644
@@ -121,7 +121,7 @@ enum {
 #define SA_AIF_PDEV_CHANGE		(1<<4)
 #define SA_AIF_LDEV_CHANGE		(1<<5)
 #define SA_AIF_BPSTAT_CHANGE		(1<<30)
-#define SA_AIF_BPCFG_CHANGE		(1<<31)
+#define SA_AIF_BPCFG_CHANGE		(1U<<31)
 
 #define HBA_MAX_SG_EMBEDDED		28
 #define HBA_MAX_SG_SEPARATE		90
Signature · 1 line
2.35.1
Reviewed-by on v1 1/11
#2 lore raw

Re: [PATCH 01/11] scsi: aacraid: Fix undefined behavior due to shift overflowing the constant

Quoted 9 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix

drivers/scsi/aacraid/commsup.c: In function ‘aac_handle_sa_aif’:
drivers/scsi/aacraid/commsup.c:1983:2: error: case label does not reduce to an integer constant

    case SA_AIF_BPCFG_CHANGE:
    ^~~~
Quoted 8 lines

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Adaptec OEM Raid Solutions <[email protected]>
Cc "James E.J. Bottomley" <[email protected]>
Cc "Martin K. Petersen" <[email protected]>

Thanks.

Quoted 17 lines
---
 drivers/scsi/aacraid/aacraid.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/aacraid/aacraid.h +1 -1
index f849e7c9d428..5e115e8b2ba4 100644
@@ -121,7 +121,7 @@ enum {
 #define SA_AIF_PDEV_CHANGE		(1<<4)
 #define SA_AIF_LDEV_CHANGE		(1<<5)
 #define SA_AIF_BPSTAT_CHANGE		(1<<30)
-#define SA_AIF_BPCFG_CHANGE		(1<<31)
+#define SA_AIF_BPCFG_CHANGE		(1U<<31)
 
 #define HBA_MAX_SG_EMBEDDED		28
 #define HBA_MAX_SG_SEPARATE		90
Signature · 1 line
~Randy
#3 lore raw

Re: [PATCH 01/11] scsi: aacraid: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

Borislav,

Quoted 4 lines

drivers/scsi/aacraid/commsup.c: In function ‘aac_handle_sa_aif’:
drivers/scsi/aacraid/commsup.c:1983:2: error: case label does not reduce to an integer constant

    case SA_AIF_BPCFG_CHANGE:
    ^~~~

Applied to 5.19/scsi-staging, thanks!

Signature · 1 line
Martin K. Petersen	Oracle Linux Engineering
#4 lore raw

Re: [PATCH 01/11] scsi: aacraid: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov
Quoted 10 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix

drivers/scsi/aacraid/commsup.c: In function ‘aac_handle_sa_aif’:
drivers/scsi/aacraid/commsup.c:1983:2: error: case label does not reduce to an integer constant

    case SA_AIF_BPCFG_CHANGE:
    ^~~~

[...]

Applied to 5.19/scsi-queue, thanks!

[01/11] scsi: aacraid: Fix undefined behavior due to shift overflowing the constant

        https://git.kernel.org/mkp/scsi/c/331c6e910f1a
Signature · 1 line
Martin K. Petersen	Oracle Linux Engineering
#5 lore raw

PATCH 2/11 ALSA: usb-audio: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

sound/usb/midi.c: In function ‘snd_usbmidi_out_endpoint_create’:
sound/usb/midi.c:1389:2: error: case label does not reduce to an integer constant

    case USB_ID(0xfc08, 0x0101): /* Unknown vendor Cable */
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

---
 sound/usb/usbaudio.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/usb/usbaudio.h +1 -1
index 167834133b9b..2b3dd55e8ee0 100644
@@ -8,7 +8,7 @@
  */
 
 /* handling of USB vendor/product ID pairs as 32-bit numbers */
-#define USB_ID(vendor, product) (((vendor) << 16) | (product))
+#define USB_ID(vendor, product) ((((unsigned int)vendor) << 16) | (product))
 #define USB_ID_VENDOR(id) ((id) >> 16)
 #define USB_ID_PRODUCT(id) ((u16)(id))
 
Signature · 1 line
2.35.1
#6 lore raw

Re: [PATCH 02/11] ALSA: usb-audio: Fix undefined behavior due to shift overflowing the constant

Quoted 26 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

sound/usb/midi.c: In function ‘snd_usbmidi_out_endpoint_create’:
sound/usb/midi.c:1389:2: error: case label does not reduce to an integer constant

    case USB_ID(0xfc08, 0x0101): /* Unknown vendor Cable */
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

---
 sound/usb/usbaudio.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/usb/usbaudio.h +0 -0
index 167834133b9b..2b3dd55e8ee0 100644

sound/usb/usbaudio.h line 11

Quoted 5 lines
sound/usb/usbaudio.h +1 -1
@@ -8,7 +8,7 @@
  */
 
 /* handling of USB vendor/product ID pairs as 32-bit numbers */
-#define USB_ID(vendor, product) (((vendor) << 16) | (product))
+#define USB_ID(vendor, product) ((((unsigned int)vendor) << 16) | (product))

Parentheses are needed around vendor (as usual for a macro). Could you resubmit with it?

Thanks.

Takashi

#7 lore raw

Re: [PATCH 02/11] ALSA: usb-audio: Fix undefined behavior due to shift overflowing the constant

Quoted 4 lines · Takashi Iwai
Quoted 1 line

+#define USB_ID(vendor, product) ((((unsigned int)vendor) << 16) | (product))

Parentheses are needed around vendor (as usual for a macro). Could you resubmit with it?

Or you can fix it up while applying. :)

Thx.

Signature · 4 lines
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
#8 lore raw

Re: [PATCH 02/11] ALSA: usb-audio: Fix undefined behavior due to shift overflowing the constant

Quoted 8 lines · Borislav Petkov
Quoted 4 lines · Takashi Iwai
Quoted 1 line

+#define USB_ID(vendor, product) ((((unsigned int)vendor) << 16) | (product))

Parentheses are needed around vendor (as usual for a macro). Could you resubmit with it?

Or you can fix it up while applying. :)

If you prefer, sure.

Takashi

#9 lore raw

PATCH 3/11 bnx2x: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c: In function ‘bnx2x_check_blocks_with_parity3’:
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c:4917:4: error: case label does not reduce to an integer constant

      case AEU_INPUTS_ATTN_BITS_MCP_LATCHED_SCPAD_PARITY:
      ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Ariel Elior <[email protected]>
Cc Sudarsana Kalluru <[email protected]>
Cc Manish Chopra <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jakub Kicinski <[email protected]>
Cc Paolo Abeni <[email protected]>
---
 drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h +1 -1
index 5caa75b41b73..881ac33fe914 100644
@@ -6218,7 +6218,7 @@
 #define AEU_INPUTS_ATTN_BITS_GPIO0_FUNCTION_0			 (0x1<<2)
 #define AEU_INPUTS_ATTN_BITS_IGU_PARITY_ERROR			 (0x1<<12)
 #define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_ROM_PARITY		 (0x1<<28)
-#define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_SCPAD_PARITY		 (0x1<<31)
+#define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_SCPAD_PARITY		 (0x1U<<31)
 #define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_UMP_RX_PARITY		 (0x1<<29)
 #define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_UMP_TX_PARITY		 (0x1<<30)
 #define AEU_INPUTS_ATTN_BITS_MISC_HW_INTERRUPT			 (0x1<<15)
Signature · 1 line
2.35.1
maintainer
#10 lore raw

Re: [PATCH 03/11] bnx2x: Fix undefined behavior due to shift overflowing the constant

Quoted 6 lines · On Tue, 5 Apr 2022 17:15:09 +0200 Borislav Petkov wrote:

From: Borislav Petkov <[email protected]>

Fix:

drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c: In function ‘bnx2x_check_blocks_with_parity3’:
drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c:4917:4: error: case label does not reduce to an integer constant

Quoted 7 lines
      case AEU_INPUTS_ATTN_BITS_MCP_LATCHED_SCPAD_PARITY:
      ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

I think this patch did not make it to netdev patchwork. Could you resend (as a non-series patch - drop the 03/11 from the subject, that way build bot will not consider it a partial/broken posting)? Thanks!

Quoted 13 lines
drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h +1 -1
index 5caa75b41b73..881ac33fe914 100644
@@ -6218,7 +6218,7 @@
 #define AEU_INPUTS_ATTN_BITS_GPIO0_FUNCTION_0			 (0x1<<2)
 #define AEU_INPUTS_ATTN_BITS_IGU_PARITY_ERROR			 (0x1<<12)
 #define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_ROM_PARITY		 (0x1<<28)
-#define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_SCPAD_PARITY		 (0x1<<31)
+#define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_SCPAD_PARITY		 (0x1U<<31)
 #define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_UMP_RX_PARITY		 (0x1<<29)
 #define AEU_INPUTS_ATTN_BITS_MCP_LATCHED_UMP_TX_PARITY		 (0x1<<30)
 #define AEU_INPUTS_ATTN_BITS_MISC_HW_INTERRUPT			 (0x1<<15)
#11 lore raw

Re: [PATCH 03/11] bnx2x: Fix undefined behavior due to shift overflowing the constant

Quoted 4 lines · Jakub Kicinski

I think this patch did not make it to netdev patchwork. Could you resend (as a non-series patch - drop the 03/11 from the subject, that way build bot will not consider it a partial/broken posting)? Thanks!

Yeah, will give vger some time as it sounds like it is clogged at the moment. Which would explain why my patches haven't appeared in multiple patchworks, if they depend on vger, that is...

I'll do what you suggest tomorrow if it doesn't appear by then.

Thx.

Signature · 4 lines
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
#12 lore raw

PATCH 4/11 drm/r128: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/gpu/drm/r128/r128_cce.c: In function ‘r128_do_init_cce’:
drivers/gpu/drm/r128/r128_cce.c:417:2: error: case label does not reduce to an integer constant

    case R128_PM4_64BM_64VCBM_64INDBM:
    ^~~~
  drivers/gpu/drm/r128/r128_cce.c:418:2: error: case label does not reduce to an integer constant
    case R128_PM4_64PIO_64VCPIO_64INDPIO:
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc David Airlie <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Alex Deucher <[email protected]>
---
 drivers/gpu/drm/r128/r128_drv.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/r128/r128_drv.h +2 -2
index 2e1bc01aa5c9..970e192b0d51 100644
@@ -300,8 +300,8 @@ extern long r128_compat_ioctl(struct file *filp, unsigned int cmd,
 #	define R128_PM4_64PIO_128INDBM		(5  << 28)
 #	define R128_PM4_64BM_128INDBM		(6  << 28)
 #	define R128_PM4_64PIO_64VCBM_64INDBM	(7  << 28)
-#	define R128_PM4_64BM_64VCBM_64INDBM	(8  << 28)
-#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15 << 28)
+#	define R128_PM4_64BM_64VCBM_64INDBM	(8U  << 28)
+#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15U << 28)
 #	define R128_PM4_BUFFER_CNTL_NOUPDATE	(1  << 27)
 
 #define R128_PM4_BUFFER_WM_CNTL		0x0708
Signature · 1 line
2.35.1
Reviewed-by on v1 4/11
#13 lore raw

Re: [PATCH 04/11] drm/r128: Fix undefined behavior due to shift overflowing the constant

Quoted 12 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/gpu/drm/r128/r128_cce.c: In function ‘r128_do_init_cce’:
drivers/gpu/drm/r128/r128_cce.c:417:2: error: case label does not reduce to an integer constant

    case R128_PM4_64BM_64VCBM_64INDBM:
    ^~~~
  drivers/gpu/drm/r128/r128_cce.c:418:2: error: case label does not reduce to an integer constant
    case R128_PM4_64PIO_64VCPIO_64INDPIO:
    ^~~~
Quoted 8 lines

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc David Airlie <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Alex Deucher <[email protected]>

Thanks.

Quoted 19 lines
---
 drivers/gpu/drm/r128/r128_drv.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/r128/r128_drv.h +2 -2
index 2e1bc01aa5c9..970e192b0d51 100644
@@ -300,8 +300,8 @@ extern long r128_compat_ioctl(struct file *filp, unsigned int cmd,
 #	define R128_PM4_64PIO_128INDBM		(5  << 28)
 #	define R128_PM4_64BM_128INDBM		(6  << 28)
 #	define R128_PM4_64PIO_64VCBM_64INDBM	(7  << 28)
-#	define R128_PM4_64BM_64VCBM_64INDBM	(8  << 28)
-#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15 << 28)
+#	define R128_PM4_64BM_64VCBM_64INDBM	(8U  << 28)
+#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15U << 28)
 #	define R128_PM4_BUFFER_CNTL_NOUPDATE	(1  << 27)
 
 #define R128_PM4_BUFFER_WM_CNTL		0x0708
Signature · 1 line
~Randy
#14 lore raw

Re: [PATCH 04/11] drm/r128: Fix undefined behavior due to shift overflowing the constant

Quoted 18 lines · Randy Dunlap
Quoted 15 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/gpu/drm/r128/r128_cce.c: In function ‘r128_do_init_cce’:
drivers/gpu/drm/r128/r128_cce.c:417:2: error: case label does not reduce to an integer constant

    case R128_PM4_64BM_64VCBM_64INDBM:
    ^~~~
  drivers/gpu/drm/r128/r128_cce.c:418:2: error: case label does not reduce to an integer constant
    case R128_PM4_64PIO_64VCPIO_64INDPIO:
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Quoted 7 lines
Quoted 5 lines
Cc David Airlie <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Alex Deucher <[email protected]>

Pushed to drm-misc-next, thanks for patch&review.
-Daniel

Quoted 25 lines

Thanks.

Quoted 19 lines
---
 drivers/gpu/drm/r128/r128_drv.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/r128/r128_drv.h +2 -2
index 2e1bc01aa5c9..970e192b0d51 100644
@@ -300,8 +300,8 @@ extern long r128_compat_ioctl(struct file *filp, unsigned int cmd,
 #	define R128_PM4_64PIO_128INDBM		(5  << 28)
 #	define R128_PM4_64BM_128INDBM		(6  << 28)
 #	define R128_PM4_64PIO_64VCBM_64INDBM	(7  << 28)
-#	define R128_PM4_64BM_64VCBM_64INDBM	(8  << 28)
-#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15 << 28)
+#	define R128_PM4_64BM_64VCBM_64INDBM	(8U  << 28)
+#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15U << 28)
 #	define R128_PM4_BUFFER_CNTL_NOUPDATE	(1  << 27)
 
 #define R128_PM4_BUFFER_WM_CNTL		0x0708
Signature · 1 line
~Randy
Signature · 3 lines
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
#15 lore raw

Re: [PATCH 04/11] drm/r128: Fix undefined behavior due to shift overflowing the constant

Quoted 25 lines · Daniel Vetter
Quoted 24 lines · Randy Dunlap
Quoted 20 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/gpu/drm/r128/r128_cce.c: In function ‘r128_do_init_cce’:
drivers/gpu/drm/r128/r128_cce.c:417:2: error: case label does not reduce to an integer constant

    case R128_PM4_64BM_64VCBM_64INDBM:
    ^~~~
  drivers/gpu/drm/r128/r128_cce.c:418:2: error: case label does not reduce to an integer constant
    case R128_PM4_64PIO_64VCPIO_64INDPIO:
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc David Airlie <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Alex Deucher <[email protected]>
Quoted 5 lines
Quoted 1 line

Pushed to drm-misc-next, thanks for patch&review.
-Daniel

Hi,

Will this be merged to mainline any time soonish?

thanks.

Quoted 26 lines
Quoted 25 lines

Thanks.

Quoted 19 lines
---
 drivers/gpu/drm/r128/r128_drv.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/r128/r128_drv.h +2 -2
index 2e1bc01aa5c9..970e192b0d51 100644
@@ -300,8 +300,8 @@ extern long r128_compat_ioctl(struct file *filp, unsigned int cmd,
 #	define R128_PM4_64PIO_128INDBM		(5  << 28)
 #	define R128_PM4_64BM_128INDBM		(6  << 28)
 #	define R128_PM4_64PIO_64VCBM_64INDBM	(7  << 28)
-#	define R128_PM4_64BM_64VCBM_64INDBM	(8  << 28)
-#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15 << 28)
+#	define R128_PM4_64BM_64VCBM_64INDBM	(8U  << 28)
+#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15U << 28)
 #	define R128_PM4_BUFFER_CNTL_NOUPDATE	(1  << 27)
 
 #define R128_PM4_BUFFER_WM_CNTL		0x0708
Signature · 1 line
~Randy
Signature · 1 line
~Randy
#16 lore raw

Re: [PATCH 04/11] drm/r128: Fix undefined behavior due to shift overflowing the constant

Quoted 30 lines · Randy Dunlap
Quoted 27 lines · Daniel Vetter
Quoted 25 lines · Randy Dunlap
Quoted 12 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/gpu/drm/r128/r128_cce.c: In function ‘r128_do_init_cce’:
drivers/gpu/drm/r128/r128_cce.c:417:2: error: case label does not reduce to an integer constant

    case R128_PM4_64BM_64VCBM_64INDBM:
    ^~~~
  drivers/gpu/drm/r128/r128_cce.c:418:2: error: case label does not reduce to an integer constant
    case R128_PM4_64PIO_64VCPIO_64INDPIO:
    ^~~~
Quoted 8 lines

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc David Airlie <[email protected]>
Cc Daniel Vetter <[email protected]>
Cc Alex Deucher <[email protected]>
Quoted 7 lines
Quoted 3 lines

Pushed to drm-misc-next, thanks for patch&review.
-Daniel

Hi,

Will this be merged to mainline any time soonish?

It missed the merge window by a hair, so it's in linux-next and will get into the next one.
-Daniel

Quoted 32 lines

thanks.

Quoted 26 lines
Quoted 25 lines

Thanks.

Quoted 19 lines
---
 drivers/gpu/drm/r128/r128_drv.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/r128/r128_drv.h +2 -2
index 2e1bc01aa5c9..970e192b0d51 100644
@@ -300,8 +300,8 @@ extern long r128_compat_ioctl(struct file *filp, unsigned int cmd,
 #	define R128_PM4_64PIO_128INDBM		(5  << 28)
 #	define R128_PM4_64BM_128INDBM		(6  << 28)
 #	define R128_PM4_64PIO_64VCBM_64INDBM	(7  << 28)
-#	define R128_PM4_64BM_64VCBM_64INDBM	(8  << 28)
-#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15 << 28)
+#	define R128_PM4_64BM_64VCBM_64INDBM	(8U  << 28)
+#	define R128_PM4_64PIO_64VCPIO_64INDPIO	(15U << 28)
 #	define R128_PM4_BUFFER_CNTL_NOUPDATE	(1  << 27)
 
 #define R128_PM4_BUFFER_WM_CNTL		0x0708
Signature · 1 line
~Randy
Signature · 1 line
~Randy
Signature · 3 lines
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
#17 lore raw

PATCH 5/11 i2c: ismt: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/i2c/busses/i2c-ismt.c: In function ‘ismt_hw_init’:
drivers/i2c/busses/i2c-ismt.c:770:2: error: case label does not reduce to an integer constant

    case ISMT_SPGT_SPD_400K:
    ^~~~
  drivers/i2c/busses/i2c-ismt.c:773:2: error: case label does not reduce to an integer constant
    case ISMT_SPGT_SPD_1M:
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

---
 drivers/i2c/busses/i2c-ismt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/i2c/busses/i2c-ismt.c +2 -2
index f4820fd3dc13..c0364314877e 100644
@@ -145,8 +145,8 @@
 #define ISMT_SPGT_SPD_MASK	0xc0000000	/* SMBus Speed mask */
 #define ISMT_SPGT_SPD_80K	0x00		/* 80 kHz */
 #define ISMT_SPGT_SPD_100K	(0x1 << 30)	/* 100 kHz */
-#define ISMT_SPGT_SPD_400K	(0x2 << 30)	/* 400 kHz */
-#define ISMT_SPGT_SPD_1M	(0x3 << 30)	/* 1 MHz */
+#define ISMT_SPGT_SPD_400K	(0x2U << 30)	/* 400 kHz */
+#define ISMT_SPGT_SPD_1M	(0x3U << 30)	/* 1 MHz */
 
 
 /* MSI Control Register (MSICTL) bit definitions */
Signature · 1 line
2.35.1
Reviewed-by on v1 5/11
#18 lore raw

Re: [PATCH 05/11] i2c: ismt: Fix undefined behavior due to shift overflowing the constant

Quoted 14 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/i2c/busses/i2c-ismt.c: In function ‘ismt_hw_init’:
drivers/i2c/busses/i2c-ismt.c:770:2: error: case label does not
reduce to an integer constant

    case ISMT_SPGT_SPD_400K:
    ^~~~
  drivers/i2c/busses/i2c-ismt.c:773:2: error: case label does not
reduce to an integer constant
    case ISMT_SPGT_SPD_1M:
    ^~~~
Quoted 4 lines

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Quoted 25 lines
Cc Seth Heasley <[email protected]>
Cc Neil Horman <[email protected]>
Cc Wolfram Sang <[email protected]>
---
 drivers/i2c/busses/i2c-ismt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/i2c/busses/i2c- renamed from drivers/i2c/busses/i2c-ismt.c +0 -0

ismt.c
index f4820fd3dc13..c0364314877e 100644

drivers/i2c/busses/i2c-ismt.c +2 -2
@@ -145,8 +145,8 @@
 #define ISMT_SPGT_SPD_MASK	0xc0000000	/* SMBus Speed mask
 */
 #define ISMT_SPGT_SPD_80K	0x00		/* 80 kHz */
 #define ISMT_SPGT_SPD_100K	(0x1 << 30)	/* 100 kHz */
-#define ISMT_SPGT_SPD_400K	(0x2 << 30)	/* 400 kHz */
-#define ISMT_SPGT_SPD_1M	(0x3 << 30)	/* 1 MHz */
+#define ISMT_SPGT_SPD_400K	(0x2U << 30)	/* 400 kHz */
+#define ISMT_SPGT_SPD_1M	(0x3U << 30)	/* 1 MHz */
 
 

/* MSI Control Register (MSICTL) bit definitions */

#19 lore raw

PATCH 6/11 brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c: In function ‘brcmf_sdio_drivestrengthinit’:
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3798:2: error: case label does not reduce to an integer constant

    case SDIOD_DRVSTR_KEY(BRCM_CC_43143_CHIP_ID, 17):
    ^~~~
  drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3809:2: error: case label does not reduce to an integer constant
    case SDIOD_DRVSTR_KEY(BRCM_CC_43362_CHIP_ID, 13):
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Arend van Spriel <[email protected]>
Cc Franky Lin <[email protected]>
Cc Hante Meuleman <[email protected]>
Cc Kalle Valo <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jakub Kicinski <[email protected]>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +1 -1
index ba3c159111d3..d78ccc223709 100644
@@ -557,7 +557,7 @@ enum brcmf_sdio_frmtype {
 	BRCMF_SDIO_FT_SUB,
 };
 
-#define SDIOD_DRVSTR_KEY(chip, pmu)     (((chip) << 16) | (pmu))
+#define SDIOD_DRVSTR_KEY(chip, pmu)     (((unsigned int)(chip) << 16) | (pmu))
 
 /* SDIO Pad drive strength to select value mappings */
 struct sdiod_drive_str {
Signature · 1 line
2.35.1
Acked-by on v1 6/11
#20 lore raw

Re: [PATCH 06/11] brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant

+ linux-wireless

Quoted 33 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c: In function ‘brcmf_sdio_drivestrengthinit’:
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3798:2: error: case label does not reduce to an integer constant

    case SDIOD_DRVSTR_KEY(BRCM_CC_43143_CHIP_ID, 17):
    ^~~~
  drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3809:2: error: case label does not reduce to an integer constant
    case SDIOD_DRVSTR_KEY(BRCM_CC_43362_CHIP_ID, 13):
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Arend van Spriel <[email protected]>
Cc Franky Lin <[email protected]>
Cc Hante Meuleman <[email protected]>
Cc Kalle Valo <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jakub Kicinski <[email protected]>
---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +0 -0
index ba3c159111d3..d78ccc223709 100644

drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c enum brcmf_sdio_frmtype { · line 560

Quoted 5 lines
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +1 -1
@@ -557,7 +557,7 @@ enum brcmf_sdio_frmtype {
 	BRCMF_SDIO_FT_SUB,
 };
 
-#define SDIOD_DRVSTR_KEY(chip, pmu)     (((chip) << 16) | (pmu))
+#define SDIOD_DRVSTR_KEY(chip, pmu)     (((unsigned int)(chip) << 16) | (pmu))

Via which tree is this going? I assume not the wireless tree, so:

Signature · 3 lines
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
#21 lore raw

Re: [PATCH 06/11] brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant

Quoted 1 line · Kalle Valo

Via which tree is this going? I assume not the wireless tree, so:

Whoever picks it up.

Quoted 1 line

Thx.

Signature · 4 lines
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
#22 lore raw

Re: [PATCH 06/11] brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant

Quoted 4 lines · Borislav Petkov
Quoted 1 line · Kalle Valo

Via which tree is this going? I assume not the wireless tree, so:

Whoever picks it up.

It would be good to have a plan so the patch is not forgotten :)

Normally brcmfmac patches go via the wireless tree, so I could take this patch. But you didn't CC linux-wireless so our patchwork doesn't see it. So if you want me to take this you need to resend.

Signature · 3 lines
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
#23 lore raw

PATCH RESEND 6/11 brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant

Fix:

drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c: In function ‘brcmf_sdio_drivestrengthinit’:
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3798:2: error: case label does not reduce to an integer constant

    case SDIOD_DRVSTR_KEY(BRCM_CC_43143_CHIP_ID, 17):
    ^~~~
  drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3809:2: error: case label does not reduce to an integer constant
    case SDIOD_DRVSTR_KEY(BRCM_CC_43362_CHIP_ID, 13):
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Arend van Spriel <[email protected]>
Cc Franky Lin <[email protected]>
Cc Hante Meuleman <[email protected]>
Cc Kalle Valo <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jakub Kicinski <[email protected]>

---

Resend, this time with linux-wireless on Cc so that patchwork can pick it up.

Thx.

 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +1 -1
index ba3c159111d3..d78ccc223709 100644
@@ -557,7 +557,7 @@ enum brcmf_sdio_frmtype {
 	BRCMF_SDIO_FT_SUB,
 };
 
-#define SDIOD_DRVSTR_KEY(chip, pmu)     (((chip) << 16) | (pmu))
+#define SDIOD_DRVSTR_KEY(chip, pmu)     (((unsigned int)(chip) << 16) | (pmu))
 
 /* SDIO Pad drive strength to select value mappings */
 struct sdiod_drive_str {
Signature · 8 lines
2.35.1


-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
maintainer Acked-by on v1 6/11
#24 lore raw

Re: [RESEND PATCH 06/11] brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant

Quoted 10 lines · Borislav Petkov

Fix:

   drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c: In function ‘brcmf_sdio_drivestrengthinit’:
   drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3798:2: error: case label does not reduce to an integer constant
     case SDIOD_DRVSTR_KEY(BRCM_CC_43143_CHIP_ID, 17):
     ^~~~
   drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3809:2: error: case label does not reduce to an integer constant
     case SDIOD_DRVSTR_KEY(BRCM_CC_43362_CHIP_ID, 13):
     ^~~~
Quoted 2 lines

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

The details aren't that gory ;-)

Quoted 9 lines
Cc Arend van Spriel <[email protected]>
Cc Franky Lin <[email protected]>
Cc Hante Meuleman <[email protected]>
Cc Kalle Valo <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jakub Kicinski <[email protected]>
#25 lore raw

Re: [RESEND PATCH 06/11] brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov
Quoted 13 lines · Borislav Petkov

Fix:

drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c: In function ‘brcmf_sdio_drivestrengthinit’:
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3798:2: error: case label does not reduce to an integer constant

    case SDIOD_DRVSTR_KEY(BRCM_CC_43143_CHIP_ID, 17):
    ^~~~
  drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3809:2: error: case label does not reduce to an integer constant
    case SDIOD_DRVSTR_KEY(BRCM_CC_43362_CHIP_ID, 13):
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Quoted 10 lines
Cc Arend van Spriel <[email protected]>
Cc Franky Lin <[email protected]>
Cc Hante Meuleman <[email protected]>
Cc Kalle Valo <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jakub Kicinski <[email protected]>

Patch applied to wireless.git, thanks.

6fb3a5868b21 brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant

Signature · 3 lines
https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
#26 lore raw

PATCH 7/11 usb: typec: tcpm: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/usb/typec/tcpm/tcpm.c: In function ‘run_state_machine’:
drivers/usb/typec/tcpm/tcpm.c:4724:3: error: case label does not reduce to an integer constant

     case BDO_MODE_TESTDATA:
     ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

---
 include/linux/usb/pd_bdo.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/usb/pd_bdo.h +1 -1
index 033fe3e17141..7c25b88d79f9 100644
@@ -15,7 +15,7 @@
 #define BDO_MODE_CARRIER2	(5 << 28)
 #define BDO_MODE_CARRIER3	(6 << 28)
 #define BDO_MODE_EYE		(7 << 28)
-#define BDO_MODE_TESTDATA	(8 << 28)
+#define BDO_MODE_TESTDATA	(8U << 28)
 
 #define BDO_MODE_MASK(mode)	((mode) & 0xf0000000)
 
Signature · 1 line
2.35.1
#27 lore raw

PATCH 8/11 mt76: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/net/wireless/mediatek/mt76/mt76x2/pci.c: In function ‘mt76x2e_probe’:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_946’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Felix Fietkau <[email protected]>
Cc Lorenzo Bianconi <[email protected]>
Cc Ryder Lee <[email protected]>
Cc Shayne Chen <[email protected]>
Cc Sean Wang <[email protected]>
Cc Kalle Valo <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jakub Kicinski <[email protected]>
---
 drivers/net/wireless/mediatek/mt76/mt76x2/pci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt76x2/pci.c +1 -1
index 8a22ee581674..df85ebc6e1df 100644
@@ -80,7 +80,7 @@ mt76x2e_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	mt76_rmw_field(dev, 0x15a10, 0x1f << 16, 0x9);
 
 	/* RG_SSUSB_G1_CDR_BIC_LTR = 0xf */
-	mt76_rmw_field(dev, 0x15a0c, 0xf << 28, 0xf);
+	mt76_rmw_field(dev, 0x15a0c, 0xfU << 28, 0xf);
 
 	/* RG_SSUSB_CDR_BR_PE1D = 0x3 */
 	mt76_rmw_field(dev, 0x15c58, 0x3 << 6, 0x3);
Signature · 1 line
2.35.1
#28 lore raw

Re: [PATCH 08/11] mt76: Fix undefined behavior due to shift overflowing the constant

Quoted 9 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/net/wireless/mediatek/mt76/mt76x2/pci.c: In function ‘mt76x2e_probe’:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_946’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)

Quoted 4 lines

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

As this fixes a compiler warning in Linus' tree, I would like to take this to wireless tree and I assigned this to myself in patchwork.

Felix, ack?

Signature · 3 lines
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
#29 lore raw

Re: [PATCH 08/11] mt76: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov
Quoted 12 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/net/wireless/mediatek/mt76/mt76x2/pci.c: In function ‘mt76x2e_probe’:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_946’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Quoted 11 lines
Cc Felix Fietkau <[email protected]>
Cc Lorenzo Bianconi <[email protected]>
Cc Ryder Lee <[email protected]>
Cc Shayne Chen <[email protected]>
Cc Sean Wang <[email protected]>
Cc Kalle Valo <[email protected]>
Cc "David S. Miller" <[email protected]>
Cc Jakub Kicinski <[email protected]>

Patch applied to wireless.git, thanks.

dbc2b1764734 mt76: Fix undefined behavior due to shift overflowing the constant

Signature · 3 lines
https://patchwork.kernel.org/project/linux-wireless/patch/[email protected]/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
#30 lore raw

PATCH 9/11 perf/imx_ddr: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

In file included from <command-line>:0:0:
In function ‘ddr_perf_counter_enable’,
inlined from ‘ddr_perf_irq_handler’ at drivers/perf/fsl_imx8_ddr_perf.c:651:2:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_729’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
...

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Frank Li <[email protected]>
Cc Will Deacon <[email protected]>
Cc Mark Rutland <[email protected]>
Cc Shawn Guo <[email protected]>
Cc Sascha Hauer <[email protected]>
Cc Pengutronix Kernel Team <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc NXP Linux Team <[email protected]>
---
 drivers/perf/fsl_imx8_ddr_perf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/perf/fsl_imx8_ddr_perf.c +1 -1
index 94ebc1ecace7..b1b2a55de77f 100644
@@ -29,7 +29,7 @@
 #define CNTL_OVER_MASK		0xFFFFFFFE
 
 #define CNTL_CSV_SHIFT		24
-#define CNTL_CSV_MASK		(0xFF << CNTL_CSV_SHIFT)
+#define CNTL_CSV_MASK		(0xFFU << CNTL_CSV_SHIFT)
 
 #define EVENT_CYCLES_ID		0
 #define EVENT_CYCLES_COUNTER	0
Signature · 7 lines
2.35.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
maintainer Acked-by on v1 9/11
#31 lore raw

Re: [PATCH 09/11] perf/imx_ddr: Fix undefined behavior due to shift overflowing the constant

Quoted 34 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

In file included from <command-line>:0:0:
In function ‘ddr_perf_counter_enable’,
inlined from ‘ddr_perf_irq_handler’ at drivers/perf/fsl_imx8_ddr_perf.c:651:2:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_729’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
...

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Frank Li <[email protected]>
Cc Will Deacon <[email protected]>
Cc Mark Rutland <[email protected]>
Cc Shawn Guo <[email protected]>
Cc Sascha Hauer <[email protected]>
Cc Pengutronix Kernel Team <[email protected]>
Cc Fabio Estevam <[email protected]>
Cc NXP Linux Team <[email protected]>
---
 drivers/perf/fsl_imx8_ddr_perf.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/perf/fsl_imx8_ddr_perf.c +0 -0
index 94ebc1ecace7..b1b2a55de77f 100644

drivers/perf/fsl_imx8_ddr_perf.c line 32

Quoted 5 lines
drivers/perf/fsl_imx8_ddr_perf.c +1 -1
@@ -29,7 +29,7 @@
 #define CNTL_OVER_MASK		0xFFFFFFFE
 
 #define CNTL_CSV_SHIFT		24
-#define CNTL_CSV_MASK		(0xFF << CNTL_CSV_SHIFT)
+#define CNTL_CSV_MASK		(0xFFU << CNTL_CSV_SHIFT)

(let me know if you'd prefer for me to queue this directly)

Will

#32 lore raw

Re: [PATCH 09/11] perf/imx_ddr: Fix undefined behavior due to shift overflowing the constant

Quoted 1 line · Will Deacon

(let me know if you'd prefer for me to queue this directly)

Yes please.

Thx.

Signature · 4 lines
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
#33 lore raw

PATCH 10/11 IB/mlx5: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/infiniband/hw/mlx5/main.c: In function ‘translate_eth_legacy_proto_oper’:
drivers/infiniband/hw/mlx5/main.c:370:2: error: case label does not reduce to an integer constant

    case MLX5E_PROT_MASK(MLX5E_50GBASE_KR2):
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

---
 include/linux/mlx5/port.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/mlx5/port.h +1 -1
index 28a928b0684b..e96ee1e348cb 100644
@@ -141,7 +141,7 @@ enum mlx5_ptys_width {
 	MLX5_PTYS_WIDTH_12X	= 1 << 4,
 };
 
-#define MLX5E_PROT_MASK(link_mode) (1 << link_mode)
+#define MLX5E_PROT_MASK(link_mode) (1U << link_mode)
 #define MLX5_GET_ETH_PROTO(reg, out, ext, field)	\
 	(ext ? MLX5_GET(reg, out, ext_##field) :	\
 	MLX5_GET(reg, out, field))
Signature · 1 line
2.35.1
#34 lore raw

Re: [PATCH 10/11] IB/mlx5: Fix undefined behavior due to shift overflowing the constant

Quoted 12 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/infiniband/hw/mlx5/main.c: In function ‘translate_eth_legacy_proto_oper’:
drivers/infiniband/hw/mlx5/main.c:370:2: error: case label does not reduce to an integer constant

    case MLX5E_PROT_MASK(MLX5E_50GBASE_KR2):
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Quoted 8 lines
---
 include/linux/mlx5/port.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

I would like to take this patch to mlx5-next, but it didn't show nor in patchworks [1] nor in lore [2].

Thanks

[1] https://patchwork.kernel.org/project/linux-rdma/list/
[2] https://lore.kernel.org/linux-rdma/

Quoted 17 lines
include/linux/mlx5/port.h +1 -1
index 28a928b0684b..e96ee1e348cb 100644
@@ -141,7 +141,7 @@ enum mlx5_ptys_width {
 	MLX5_PTYS_WIDTH_12X	= 1 << 4,
 };
 
-#define MLX5E_PROT_MASK(link_mode) (1 << link_mode)
+#define MLX5E_PROT_MASK(link_mode) (1U << link_mode)
 #define MLX5_GET_ETH_PROTO(reg, out, ext, field)	\
 	(ext ? MLX5_GET(reg, out, ext_##field) :	\
 	MLX5_GET(reg, out, field))
Signature · 1 line
2.35.1
#35 lore raw

Re: [PATCH 10/11] IB/mlx5: Fix undefined behavior due to shift overflowing the constant

Quoted 2 lines · Leon Romanovsky

I would like to take this patch to mlx5-next, but it didn't show nor in patchworks [1] nor in lore [2].

I'm investigating. I'll resend tomorrow if it hasn't appeared by then.

Thx.

Signature · 4 lines
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
#36 lore raw

Re: [PATCH 10/11] IB/mlx5: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov
Quoted 12 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

drivers/infiniband/hw/mlx5/main.c: In function ‘translate_eth_legacy_proto_oper’:
drivers/infiniband/hw/mlx5/main.c:370:2: error: case label does not reduce to an integer constant

    case MLX5E_PROT_MASK(MLX5E_50GBASE_KR2):
    ^~~~

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Quoted 8 lines
---
 include/linux/mlx5/port.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Thanks, applied to mlx5-next.

0276bd3a94c0 ("IB/mlx5: Fix undefined behavior due to shift overflowing the constant")

#37 lore raw

PATCH 11/11 drm/i915: Fix undefined behavior due to shift overflowing the constant

In reply to Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

In file included from <command-line>:0:0:
drivers/gpu/drm/i915/gt/uc/intel_guc.c: In function ‘intel_guc_send_mmio’:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_1047’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)

and other build errors due to shift overflowing values.

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Cc Jani Nikula <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Cc Rodrigo Vivi <[email protected]>
Cc Tvrtko Ursulin <[email protected]>
Cc David Airlie <[email protected]>
Cc Daniel Vetter <[email protected]>
---
 .../gpu/drm/i915/gt/uc/abi/guc_actions_abi.h   |  2 +-
 .../i915/gt/uc/abi/guc_communication_ctb_abi.h |  2 +-
 .../gpu/drm/i915/gt/uc/abi/guc_messages_abi.h  |  2 +-
 drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h     |  2 +-
 drivers/gpu/drm/i915/i915_reg.h                | 18 +++++++++---------
 5 files changed, 13 insertions(+), 13 deletions(-)
drivers/gpu/drm/i915/gt/uc/abi/guc_actions_abi.h +1 -1
index 7afdadc7656f..e835f28c0020 100644
@@ -50,7 +50,7 @@
 
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_LEN		(GUC_HXG_REQUEST_MSG_MIN_LEN + 3u)
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_0_MBZ		GUC_HXG_REQUEST_MSG_0_DATA0
-#define HOST2GUC_SELF_CFG_REQUEST_MSG_1_KLV_KEY		(0xffff << 16)
+#define HOST2GUC_SELF_CFG_REQUEST_MSG_1_KLV_KEY		(0xffffU << 16)
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_1_KLV_LEN		(0xffff << 0)
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_2_VALUE32		GUC_HXG_REQUEST_MSG_n_DATAn
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_3_VALUE64		GUC_HXG_REQUEST_MSG_n_DATAn
drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h +1 -1
index c9086a600bce..df83c1cc7c7a 100644
@@ -82,7 +82,7 @@ static_assert(sizeof(struct guc_ct_buffer_desc) == 64);
 #define GUC_CTB_HDR_LEN				1u
 #define GUC_CTB_MSG_MIN_LEN			GUC_CTB_HDR_LEN
 #define GUC_CTB_MSG_MAX_LEN			256u
-#define GUC_CTB_MSG_0_FENCE			(0xffff << 16)
+#define GUC_CTB_MSG_0_FENCE			(0xffffU << 16)
 #define GUC_CTB_MSG_0_FORMAT			(0xf << 12)
 #define   GUC_CTB_FORMAT_HXG			0u
 #define GUC_CTB_MSG_0_RESERVED			(0xf << 8)
drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h +1 -1
index 29ac823acd4c..7d5ba4d97d70 100644
@@ -40,7 +40,7 @@
  */
 
 #define GUC_HXG_MSG_MIN_LEN			1u
-#define GUC_HXG_MSG_0_ORIGIN			(0x1 << 31)
+#define GUC_HXG_MSG_0_ORIGIN			(0x1U << 31)
 #define   GUC_HXG_ORIGIN_HOST			0u
 #define   GUC_HXG_ORIGIN_GUC			1u
 #define GUC_HXG_MSG_0_TYPE			(0x7 << 28)
drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h +1 -1
index 66027a42cda9..ad570fa002a6 100644
@@ -28,7 +28,7 @@
 #define   GS_MIA_HALT_REQUESTED		  (0x02 << GS_MIA_SHIFT)
 #define   GS_MIA_ISR_ENTRY		  (0x04 << GS_MIA_SHIFT)
 #define   GS_AUTH_STATUS_SHIFT		30
-#define   GS_AUTH_STATUS_MASK		  (0x03 << GS_AUTH_STATUS_SHIFT)
+#define   GS_AUTH_STATUS_MASK		  (0x03U << GS_AUTH_STATUS_SHIFT)
 #define   GS_AUTH_STATUS_BAD		  (0x01 << GS_AUTH_STATUS_SHIFT)
 #define   GS_AUTH_STATUS_GOOD		  (0x02 << GS_AUTH_STATUS_SHIFT)
 
drivers/gpu/drm/i915/i915_reg.h +9 -9
index 3c87d77d2cf6..f3ba3d0a430b 100644
@@ -7555,19 +7555,19 @@ enum skl_power_gate {
 #define  PORT_CLK_SEL_LCPLL_810		(2 << 29)
 #define  PORT_CLK_SEL_SPLL		(3 << 29)
 #define  PORT_CLK_SEL_WRPLL(pll)	(((pll) + 4) << 29)
-#define  PORT_CLK_SEL_WRPLL1		(4 << 29)
-#define  PORT_CLK_SEL_WRPLL2		(5 << 29)
-#define  PORT_CLK_SEL_NONE		(7 << 29)
-#define  PORT_CLK_SEL_MASK		(7 << 29)
+#define  PORT_CLK_SEL_WRPLL1		(4U << 29)
+#define  PORT_CLK_SEL_WRPLL2		(5U << 29)
+#define  PORT_CLK_SEL_NONE		(7U << 29)
+#define  PORT_CLK_SEL_MASK		(7U << 29)
 
 /* On ICL+ this is the same as PORT_CLK_SEL, but all bits change. */
 #define DDI_CLK_SEL(port)		PORT_CLK_SEL(port)
 #define  DDI_CLK_SEL_NONE		(0x0 << 28)
-#define  DDI_CLK_SEL_MG			(0x8 << 28)
-#define  DDI_CLK_SEL_TBT_162		(0xC << 28)
-#define  DDI_CLK_SEL_TBT_270		(0xD << 28)
-#define  DDI_CLK_SEL_TBT_540		(0xE << 28)
-#define  DDI_CLK_SEL_TBT_810		(0xF << 28)
+#define  DDI_CLK_SEL_MG			(0x8U << 28)
+#define  DDI_CLK_SEL_TBT_162		(0xCU << 28)
+#define  DDI_CLK_SEL_TBT_270		(0xDU << 28)
+#define  DDI_CLK_SEL_TBT_540		(0xEU << 28)
+#define  DDI_CLK_SEL_TBT_810		(0xFU << 28)
 #define  DDI_CLK_SEL_MASK		(0xF << 28)
 
 /* Transcoder clock selection */
Signature · 1 line
2.35.1
Acked-by Tested-by on v1 11/11
#38 lore raw

Re: [PATCH 11/11] drm/i915: Fix undefined behavior due to shift overflowing the constant

Quoted 10 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

In file included from <command-line>:0:0:
drivers/gpu/drm/i915/gt/uc/intel_guc.c: In function ‘intel_guc_send_mmio’:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_1047’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)

Quoted 5 lines

and other build errors due to shift overflowing values.

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Is this merged anywhere?
It could/should at least be in linux-next so that other people don't waste time on it.

thanks.

Quoted 102 lines
Cc Jani Nikula <[email protected]>
Cc Joonas Lahtinen <[email protected]>
Cc Rodrigo Vivi <[email protected]>
Cc Tvrtko Ursulin <[email protected]>
Cc David Airlie <[email protected]>
Cc Daniel Vetter <[email protected]>
---
 .../gpu/drm/i915/gt/uc/abi/guc_actions_abi.h   |  2 +-
 .../i915/gt/uc/abi/guc_communication_ctb_abi.h |  2 +-
 .../gpu/drm/i915/gt/uc/abi/guc_messages_abi.h  |  2 +-
 drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h     |  2 +-
 drivers/gpu/drm/i915/i915_reg.h                | 18 +++++++++---------
 5 files changed, 13 insertions(+), 13 deletions(-)
drivers/gpu/drm/i915/gt/uc/abi/guc_actions_abi.h +1 -1
index 7afdadc7656f..e835f28c0020 100644
@@ -50,7 +50,7 @@
 
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_LEN		(GUC_HXG_REQUEST_MSG_MIN_LEN + 3u)
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_0_MBZ		GUC_HXG_REQUEST_MSG_0_DATA0
-#define HOST2GUC_SELF_CFG_REQUEST_MSG_1_KLV_KEY		(0xffff << 16)
+#define HOST2GUC_SELF_CFG_REQUEST_MSG_1_KLV_KEY		(0xffffU << 16)
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_1_KLV_LEN		(0xffff << 0)
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_2_VALUE32		GUC_HXG_REQUEST_MSG_n_DATAn
 #define HOST2GUC_SELF_CFG_REQUEST_MSG_3_VALUE64		GUC_HXG_REQUEST_MSG_n_DATAn
drivers/gpu/drm/i915/gt/uc/abi/guc_communication_ctb_abi.h +1 -1
index c9086a600bce..df83c1cc7c7a 100644
@@ -82,7 +82,7 @@ static_assert(sizeof(struct guc_ct_buffer_desc) == 64);
 #define GUC_CTB_HDR_LEN				1u
 #define GUC_CTB_MSG_MIN_LEN			GUC_CTB_HDR_LEN
 #define GUC_CTB_MSG_MAX_LEN			256u
-#define GUC_CTB_MSG_0_FENCE			(0xffff << 16)
+#define GUC_CTB_MSG_0_FENCE			(0xffffU << 16)
 #define GUC_CTB_MSG_0_FORMAT			(0xf << 12)
 #define   GUC_CTB_FORMAT_HXG			0u
 #define GUC_CTB_MSG_0_RESERVED			(0xf << 8)
drivers/gpu/drm/i915/gt/uc/abi/guc_messages_abi.h +1 -1
index 29ac823acd4c..7d5ba4d97d70 100644
@@ -40,7 +40,7 @@
  */
 
 #define GUC_HXG_MSG_MIN_LEN			1u
-#define GUC_HXG_MSG_0_ORIGIN			(0x1 << 31)
+#define GUC_HXG_MSG_0_ORIGIN			(0x1U << 31)
 #define   GUC_HXG_ORIGIN_HOST			0u
 #define   GUC_HXG_ORIGIN_GUC			1u
 #define GUC_HXG_MSG_0_TYPE			(0x7 << 28)
drivers/gpu/drm/i915/gt/uc/intel_guc_reg.h +1 -1
index 66027a42cda9..ad570fa002a6 100644
@@ -28,7 +28,7 @@
 #define   GS_MIA_HALT_REQUESTED		  (0x02 << GS_MIA_SHIFT)
 #define   GS_MIA_ISR_ENTRY		  (0x04 << GS_MIA_SHIFT)
 #define   GS_AUTH_STATUS_SHIFT		30
-#define   GS_AUTH_STATUS_MASK		  (0x03 << GS_AUTH_STATUS_SHIFT)
+#define   GS_AUTH_STATUS_MASK		  (0x03U << GS_AUTH_STATUS_SHIFT)
 #define   GS_AUTH_STATUS_BAD		  (0x01 << GS_AUTH_STATUS_SHIFT)
 #define   GS_AUTH_STATUS_GOOD		  (0x02 << GS_AUTH_STATUS_SHIFT)
 
drivers/gpu/drm/i915/i915_reg.h +9 -9
index 3c87d77d2cf6..f3ba3d0a430b 100644
@@ -7555,19 +7555,19 @@ enum skl_power_gate {
 #define  PORT_CLK_SEL_LCPLL_810		(2 << 29)
 #define  PORT_CLK_SEL_SPLL		(3 << 29)
 #define  PORT_CLK_SEL_WRPLL(pll)	(((pll) + 4) << 29)
-#define  PORT_CLK_SEL_WRPLL1		(4 << 29)
-#define  PORT_CLK_SEL_WRPLL2		(5 << 29)
-#define  PORT_CLK_SEL_NONE		(7 << 29)
-#define  PORT_CLK_SEL_MASK		(7 << 29)
+#define  PORT_CLK_SEL_WRPLL1		(4U << 29)
+#define  PORT_CLK_SEL_WRPLL2		(5U << 29)
+#define  PORT_CLK_SEL_NONE		(7U << 29)
+#define  PORT_CLK_SEL_MASK		(7U << 29)
 
 /* On ICL+ this is the same as PORT_CLK_SEL, but all bits change. */
 #define DDI_CLK_SEL(port)		PORT_CLK_SEL(port)
 #define  DDI_CLK_SEL_NONE		(0x0 << 28)
-#define  DDI_CLK_SEL_MG			(0x8 << 28)
-#define  DDI_CLK_SEL_TBT_162		(0xC << 28)
-#define  DDI_CLK_SEL_TBT_270		(0xD << 28)
-#define  DDI_CLK_SEL_TBT_540		(0xE << 28)
-#define  DDI_CLK_SEL_TBT_810		(0xF << 28)
+#define  DDI_CLK_SEL_MG			(0x8U << 28)
+#define  DDI_CLK_SEL_TBT_162		(0xCU << 28)
+#define  DDI_CLK_SEL_TBT_270		(0xDU << 28)
+#define  DDI_CLK_SEL_TBT_540		(0xEU << 28)
+#define  DDI_CLK_SEL_TBT_810		(0xFU << 28)
 #define  DDI_CLK_SEL_MASK		(0xF << 28)
 
 /* Transcoder clock selection */
Signature · 1 line
~Randy
#39 lore raw

Re: [PATCH 11/11] drm/i915: Fix undefined behavior due to shift overflowing the constant

Quoted 15 lines · Randy Dunlap
Quoted 12 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

In file included from <command-line>:0:0:
drivers/gpu/drm/i915/gt/uc/intel_guc.c: In function ‘intel_guc_send_mmio’:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_1047’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)

and other build errors due to shift overflowing values.

Quoted 8 lines
Quoted 3 lines

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Is this merged anywhere?

It's state is "new" in their patchwork:

https://patchwork.freedesktop.org/patch/480756/

so I guess not yet.

Quoted 2 lines

It could/should at least be in linux-next so that other people don't waste time on it.

-ETOOMANYPATCHES I guess. :-\

Signature · 4 lines
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
maintainer
#40 lore raw

Re: [PATCH 11/11] drm/i915: Fix undefined behavior due to shift overflowing the constant

Quoted 20 lines · Borislav Petkov
Quoted 19 lines · Randy Dunlap
Quoted 15 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Fix:

In file included from <command-line>:0:0:
drivers/gpu/drm/i915/gt/uc/intel_guc.c: In function ‘intel_guc_send_mmio’:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_1047’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)

and other build errors due to shift overflowing values.

See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.

Quoted 8 lines
Quoted 4 lines

Is this merged anywhere?

It's state is "new" in their patchwork:

https://patchwork.freedesktop.org/patch/480756/

Basically we run all patches through CI before merging, and because only one patch was sent to intel-gfx, the CI just sat waiting for the rest of the series to arrive...

Anyway, didn't really like the changes in i915_reg.h, sent my version of that and the rest separately [1].

Quoted 6 lines

so I guess not yet.

Quoted 2 lines

It could/should at least be in linux-next so that other people don't waste time on it.

-ETOOMANYPATCHES I guess. :-\

Yeah, sorry about that.

BR,
Jani.

[1] https://patchwork.freedesktop.org/series/104122/

Signature · 1 line
Jani Nikula, Intel Open Source Graphics Center
#41 lore raw
In reply to Borislav Petkov

From: Borislav Petkov

Quoted 1 line

Sent: 05 April 2022 16:15

...

Quoted 3 lines

Due to some recent changes which added -fsanitize-shift to the build options of an allmodconfig, it started failing here with an old gcc because getting an overflow while shifting is undefined C99 behavior.

Annoyingly it is extremely unlikely that a 2s compliment system would ever generate anything other than the expected value - ie the value obtained by doing the calculation with unsigned values and then getting the result cast back to signed (which is IIRC implementation defined).

It isn't as though there are many 1s compliment systems out there, and probably even fewer 'sign overpunch' ones. Never mind trying to find a working C compiler for them.

What sort of crack are the C standards body and compiler people actually smoking?

David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK Registration No: 1397386 (Wales)

maintainer
#42 lore raw
In reply to Borislav Petkov
Quoted 8 lines · Borislav Petkov

From: Borislav Petkov <[email protected]>

Hi all,

so this is the result of me trying to make allmodconfig actually build here.

[...]

Applied perf/imx_ddr patch only to arm64 (for-next/fixes), thanks!

[09/11] perf/imx_ddr: Fix undefined behavior due to shift overflowing the constant

        https://git.kernel.org/arm64/c/d02b4dd84e1a

Cheers,

Signature · 5 lines
Will

https://fixes.arm64.dev
https://next.arm64.dev
https://will.arm64.dev