Skip to contentSkip to conversation

usb: gadget: u_ether: Fix host MAC address case

PATCH In review Konrad Gräfe 8 participants 21 messages Apr 25, 2023 – May 5, 2023

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 usb: gadget: u_ether: Fix host MAC address case

As the CDC-ECM specification states the host MAC address must be sent to the host as an uppercase hexadecimal string:
The Unicode character is chosen from the set of values 30h through 39h and 41h through 46h (0-9 and A-F).

However, snprintf(.., "%pm", ..) generates a lowercase MAC address string. While most host drivers are tolerant to this, UsbNcm.sys on Windows 10 is not. Instead it uses a different MAC address with all bytes set to zero including and after the first byte containing a lowercase letter. On Windows 11 Microsoft fixed it, but apparently they did not backport the fix.

This change fixes the issue by upper-casing the MAC to comply with the specification.

---
 drivers/usb/gadget/function/u_ether.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
drivers/usb/gadget/function/u_ether.c +8 -1
index 6956ad8ba8dd..250734e090fc 100644
@@ -958,6 +958,7 @@ EXPORT_SYMBOL_GPL(gether_get_host_addr);
 int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len)
 {
 	struct eth_dev *dev;
+	int i, slen;
 
 	if (len < 13)
 		return -EINVAL;
@@ -965,7 +966,13 @@ int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len)
 	dev = netdev_priv(net);
 	snprintf(host_addr, len, "%pm", dev->host_mac);
 
-	return strlen(host_addr);
+
+	slen = strlen(host_addr);
+
+	for (i = 0; i < slen; i++)
+		host_addr[i] = toupper(host_addr[i]);
+
+	return slen;
 }
 EXPORT_SYMBOL_GPL(gether_get_host_addr_cdc);
 
maintainer
#1 lore raw
Quoted 40 lines · Konrad Gräfe

As the CDC-ECM specification states the host MAC address must be sent to the host as an uppercase hexadecimal string:
The Unicode character is chosen from the set of values 30h through 39h and 41h through 46h (0-9 and A-F).

However, snprintf(.., "%pm", ..) generates a lowercase MAC address string. While most host drivers are tolerant to this, UsbNcm.sys on Windows 10 is not. Instead it uses a different MAC address with all bytes set to zero including and after the first byte containing a lowercase letter. On Windows 11 Microsoft fixed it, but apparently they did not backport the fix.

This change fixes the issue by upper-casing the MAC to comply with the specification.

---
 drivers/usb/gadget/function/u_ether.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
drivers/usb/gadget/function/u_ether.c +1 -1
index 6956ad8ba8dd..250734e090fc 100644
@@ -958,6 +958,7 @@ EXPORT_SYMBOL_GPL(gether_get_host_addr);
 int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len)
 {
 	struct eth_dev *dev;
+	int i, slen;
 
 	if (len < 13)
 		return -EINVAL;
@@ -965,7 +966,13 @@ int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len)
 	dev = netdev_priv(net);
 	snprintf(host_addr, len, "%pm", dev->host_mac);
 
-	return strlen(host_addr);

drivers/usb/gadget/function/u_ether.c int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len) · line 978

Quoted 10 lines
drivers/usb/gadget/function/u_ether.c +7 -0
+
+	slen = strlen(host_addr);
+
+	for (i = 0; i < slen; i++)
+		host_addr[i] = toupper(host_addr[i]);
+
+	return slen;
 }
 EXPORT_SYMBOL_GPL(gether_get_host_addr_cdc);
 

Hi,

This is the friendly patch-bot of Greg Kroah-Hartman. You have sent him a patch that has triggered this response. He used to manually respond to these common problems, but in order to save his sanity (he kept writing the same thing over and over, yet to different people), I was created. Hopefully you will not take offence and will fix the problem in your patch and resubmit it so that it can be accepted into the Linux kernel tree.

You are receiving this message because of the following common error(s) as indicated below:

- This looks like a new version of a previously submitted patch, but you did not list below the --- line any changes from the previous version. Please read the section entitled "The canonical patch format" in the kernel file, Documentation/process/submitting-patches.rst for what needs to be done here to properly describe this.

If you wish to discuss this problem further, or you have questions about how to resolve this issue, please feel free to respond to this email and Greg will reply once he has dug out from the pending patches received from other developers.

thanks,

greg k-h's patch email bot

#2 lore raw

PATCH 1/1 usb: gadget: u_ether: Fix host MAC address case

As the CDC-ECM specification states the host MAC address must be sent to the host as an uppercase hexadecimal string:
The Unicode character is chosen from the set of values 30h through 39h and 41h through 46h (0-9 and A-F).

However, snprintf(.., "%pm", ..) generates a lowercase MAC address string. While most host drivers are tolerant to this, UsbNcm.sys on Windows 10 is not. Instead it uses a different MAC address with all bytes set to zero including and after the first byte containing a lowercase letter. On Windows 11 Microsoft fixed it, but apparently they did not backport the fix.

This change fixes the issue by upper-casing the MAC to comply with the specification.

---
V1 -> V2: Fixed checkpatch.pl warnings

 drivers/usb/gadget/function/u_ether.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
drivers/usb/gadget/function/u_ether.c +8 -1
index 6956ad8ba8dd..250734e090fc 100644
@@ -958,6 +958,7 @@ EXPORT_SYMBOL_GPL(gether_get_host_addr);
 int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len)
 {
 	struct eth_dev *dev;
+	int i, slen;
 
 	if (len < 13)
 		return -EINVAL;
@@ -965,7 +966,13 @@ int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len)
 	dev = netdev_priv(net);
 	snprintf(host_addr, len, "%pm", dev->host_mac);
 
-	return strlen(host_addr);
+
+	slen = strlen(host_addr);
+
+	for (i = 0; i < slen; i++)
+		host_addr[i] = toupper(host_addr[i]);
+
+	return slen;
 }
 EXPORT_SYMBOL_GPL(gether_get_host_addr_cdc);
 
#3 lore raw

Hi Konrad,

Quoted 13 lines · Konrad Gräfe

As the CDC-ECM specification states the host MAC address must be sent to the host as an uppercase hexadecimal string:
    The Unicode character is chosen from the set of values 30h through 39h and 41h through 46h (0-9 and A-F).

However, snprintf(.., "%pm", ..) generates a lowercase MAC address string. While most host drivers are tolerant to this, UsbNcm.sys on Windows 10 is not. Instead it uses a different MAC address with all bytes set to zero including and after the first byte containing a lowercase letter. On Windows 11 Microsoft fixed it, but apparently they did not backport the fix.

Quoted 7 lines

This change fixes the issue by upper-casing the MAC to comply with the specification.

---
V1 -> V2: Fixed checkpatch.pl warnings

When sending a new version, please use the -v option of git format-patch to add the v2 prefix in the [PATCH] header, see https://lore.kernel.org/lkml/[email protected]/ (here a v7).

c.f. https://www.kernel.org/doc/html/latest/process/submitting-patches.html#the-canonical-patch-format

Additionally, please use scripts/get_maintainer.pl on your patch to know who to put in the recipient list, here you're missing [email protected] which is always added. If I'm not mistaken, it's basically used for archival purposes but also so people don't have to subscribe or search through all mailing list to know what's going on globally in the kernel community.

Quoted 3 lines
  drivers/usb/gadget/function/u_ether.c | 9 ++++++++-
  1 file changed, 8 insertions(+), 1 deletion(-)

The patch should be the mail body and not attached, git send-email should be able to get you through the hoops. I've seen https://git-send-email.io/ mentioned in various places so maybe that could help you get started. See https://lore.kernel.org/lkml/[email protected]/ and yours: https://lore.kernel.org/all/[email protected]/

In yours, you can see
"""
[-- Attachment #1.1.2: 0001-usb-gadget-u_ether-Fix-host-MAC-address-case.patch --] [-- Type: text/x-patch, Size: 812 bytes --]
"""
this is not good.

The point of the mailing list patch submission process is to allow in-line review, people will typically comment in-line directly, see https://lore.kernel.org/lkml/20230426111358.xh3gbhlvxj46ggi5@CAB-WSD-L081021/ for example. Adding the patch as an attachment doesn't allow this review process to happen. There also are a few robots applying patches directly from the various mailing lists and running some build tests automatically, I am not sure they are developed with patch as an attachment in mind.

c.f. https://www.kernel.org/doc/html/latest/process/submitting-patches.html#no-mime-no-links-no-compression-no-attachments-just-plain-text

For the "i'm not so sure" part of my mail now.
I don't know if this matches stable backport policy but I believe it could/should. So maybe add:
"""
Cc: [email protected]
"""
before your Signed-off-by, c.f. https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html

I think you should add:
"""

Fixes bcd4a1c40bee ("usb: gadget: u_ether: construct with default values and add setters/getters")

"""
since (I believe) this is fixing an issue introduced in that commit.
c.f. https://www.kernel.org/doc/html/latest/process/submitting-patches.html#using-reported-by-tested-by-reviewed-by-suggested-by-and-fixes and https://www.kernel.org/doc/html/latest/process/submitting-patches.html#describe-your-changes

Cheers,
Quentin

maintainer
#4 lore raw
In reply to Konrad Gräfe
Quoted 13 lines · Konrad Gräfe

As the CDC-ECM specification states the host MAC address must be sent to the host as an uppercase hexadecimal string:
The Unicode character is chosen from the set of values 30h through 39h and 41h through 46h (0-9 and A-F).

However, snprintf(.., "%pm", ..) generates a lowercase MAC address string. While most host drivers are tolerant to this, UsbNcm.sys on Windows 10 is not. Instead it uses a different MAC address with all bytes set to zero including and after the first byte containing a lowercase letter. On Windows 11 Microsoft fixed it, but apparently they did not backport the fix.

Quoted 6 lines

This change fixes the issue by upper-casing the MAC to comply with the specification.

---
V1 -> V2: Fixed checkpatch.pl warnings

There is no "v2" in the subject line, so our tools will get confused and have no idea this is a newer patch.

Please fix up and send a v3?

Quoted 1 line

snprintf(host_addr, len, "%pm", dev->host_mac);

Is there no option to print a mac address with all uppercase? If not, why not add that instead as it's needed here and maybe other places, right?

thanks,

greg k-h

#5 lore raw

PATCH v3 1/2 vsprintf: Add %p[mM]U for uppercase MAC address

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

---
Added in v3

 lib/vsprintf.c | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)
lib/vsprintf.c +15 -3
index be71a03c936a..8aee1caabd9e 100644
@@ -1269,9 +1269,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 {
 	char mac_addr[sizeof("xx:xx:xx:xx:xx:xx")];
 	char *p = mac_addr;
-	int i;
+	int i, pos;
 	char separator;
 	bool reversed = false;
+	bool uppercase = false;
 
 	if (check_pointer(&buf, end, addr, spec))
 		return buf;
@@ -1281,6 +1282,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 		separator = '-';
 		break;
 
+	case 'U':
+		uppercase = true;
+		break;
+
 	case 'R':
 		reversed = true;
 		fallthrough;
@@ -1292,9 +1297,14 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 
 	for (i = 0; i < 6; i++) {
 		if (reversed)
-			p = hex_byte_pack(p, addr[5 - i]);
+			pos = 5 - i;
+		else
+			pos = i;
+
+		if (uppercase)
+			p = hex_byte_pack_upper(p, addr[pos]);
 		else
-			p = hex_byte_pack(p, addr[i]);
+			p = hex_byte_pack(p, addr[pos]);
 
 		if (fmt[0] == 'M' && i != 5)
 			*p++ = separator;
@@ -2279,6 +2289,7 @@ char *rust_fmt_argument(char *buf, char *end, void *ptr);
  * - 'm' For a 6-byte MAC address, it prints the hex address without colons
  * - 'MF' For a 6-byte MAC FDDI address, it prints the address
  *       with a dash-separated hex notation
+ * - '[mM]U' For a 6-byte MAC address in uppercase hex
  * - '[mM]R' For a 6-byte MAC address, Reverse order (Bluetooth)
  * - 'I' [46] for IPv4/IPv6 addresses printed in the usual way
  *       IPv4 uses dot-separated decimal without leading 0's (1.2.3.4)
@@ -2407,6 +2418,7 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr,
 	case 'M':			/* Colon separated: 00:01:02:03:04:05 */
 	case 'm':			/* Contiguous: 000102030405 */
 					/* [mM]F (FDDI) */
+					/* [mM]U (Uppercase hex) */
 					/* [mM]R (Reverse order; Bluetooth) */
 		return mac_address_string(buf, end, ptr, spec, fmt);
 	case 'I':			/* Formatted IP supported
Signature · 1 line
2.34.1
#6 lore raw

PATCH v3 2/2 usb: gadget: u_ether: Fix host MAC address case

The CDC-ECM specification [1] requires to send the host MAC address as an uppercase hexadecimal string in chapter "5.4 Ethernet Networking Functional Descriptor":
The Unicode character is chosen from the set of values 30h through 39h and 41h through 46h (0-9 and A-F).

However, snprintf(.., "%pm", ..) generates a lowercase MAC address string. While most host drivers are tolerant to this, UsbNcm.sys on Windows 10 is not. Instead it uses a different MAC address with all bytes set to zero including and after the first byte containing a lowercase letter. On Windows 11 Microsoft fixed it, but apparently they did not backport the fix.

This change fixes the issue by using "%pmU" to generate an uppercase hex string to comply with the specification.

[1]: https://www.usb.org/document-library/class-definitions-communication-devices-12, file ECM120.pdf

Fixes bcd4a1c40bee ("usb: gadget: u_ether: construct with default values and add setters/getters")

---
Changes since v2:
* Add uppercase MAC address format string and use that instead of manually uppercasing the resulting MAC address string.

Changes since v1:
* Fixed checkpatch.pl warnings

 drivers/usb/gadget/function/u_ether.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/gadget/function/u_ether.c +1 -1
index 6956ad8ba8dd..70e6b825654c 100644
@@ -963,7 +963,7 @@ int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len)
 		return -EINVAL;
 
 	dev = netdev_priv(net);
-	snprintf(host_addr, len, "%pm", dev->host_mac);
+	snprintf(host_addr, len, "%pmU", dev->host_mac);
 
 	return strlen(host_addr);
 }
Signature · 1 line
2.34.1
#7 lore raw

PATCH v4 1/2 vsprintf: Add %p[mM]U for uppercase MAC address

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

---

Changes since v3:
* Added documentation
* Added test cases
* Use string_upper() after conversion to simplify conversion loop
* Fixed maybe-uninitalized variable warning

Added in v3

 Documentation/core-api/printk-formats.rst | 15 ++++++++++-----
 lib/test_printf.c                         |  2 ++
 lib/vsprintf.c                            |  5 +++++
 3 files changed, 17 insertions(+), 5 deletions(-)
Documentation/core-api/printk-formats.rst +10 -5
index dbe1aacc79d0..1ec682bdfe94 100644
@@ -298,11 +298,13 @@ MAC/FDDI addresses
 
 ::
 
-	%pM	00:01:02:03:04:05
-	%pMR	05:04:03:02:01:00
-	%pMF	00-01-02-03-04-05
-	%pm	000102030405
-	%pmR	050403020100
+	%pM	00:01:02:03:aa:bb
+	%pMR	aa:bb:03:02:01:00
+	%pMF	00-01-02-03-aa-bb
+	%pMU	00:01:02:03:AA:BB
+	%pm	00010203aabb
+	%pmR	bbaa03020100
+	%pmU	00010203AABB
 
 For printing 6-byte MAC/FDDI addresses in hex notation. The ``M`` and ``m``
 specifiers result in a printed address with (M) or without (m) byte
@@ -316,6 +318,9 @@ For Bluetooth addresses the ``R`` specifier shall be used after the ``M``
 specifier to use reversed byte order suitable for visual interpretation
 of Bluetooth addresses which are in the little endian order.
 
+For uppercase hex notation the ``U`` specifier shall be used after the ``M``
+and ``m`` specifiers.
+
 Passed by reference.
 
 IPv4 addresses
lib/test_printf.c +2 -0
index 46b4e6c414a3..7f4de2ecafbc 100644
@@ -416,9 +416,11 @@ mac(void)
 	const u8 addr[6] = {0x2d, 0x48, 0xd6, 0xfc, 0x7a, 0x05};
 
 	test("2d:48:d6:fc:7a:05", "%pM", addr);
+	test("2D:48:D6:FC:7A:05", "%pMU", addr);
 	test("05:7a:fc:d6:48:2d", "%pMR", addr);
 	test("2d-48-d6-fc-7a-05", "%pMF", addr);
 	test("2d48d6fc7a05", "%pm", addr);
+	test("2D48D6FC7A05", "%pmU", addr);
 	test("057afcd6482d", "%pmR", addr);
 }
 
lib/vsprintf.c +5 -0
index be71a03c936a..c82616c335e0 100644
@@ -1301,6 +1301,9 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 	}
 	*p = '\0';
 
+	if (fmt[1] == 'U')
+		string_upper(mac_addr, mac_addr);
+
 	return string_nocheck(buf, end, mac_addr, spec);
 }
 
@@ -2280,6 +2283,7 @@ char *rust_fmt_argument(char *buf, char *end, void *ptr);
  * - 'MF' For a 6-byte MAC FDDI address, it prints the address
  *       with a dash-separated hex notation
  * - '[mM]R' For a 6-byte MAC address, Reverse order (Bluetooth)
+ * - '[mM]U' For a 6-byte MAC address in uppercase hex
  * - 'I' [46] for IPv4/IPv6 addresses printed in the usual way
  *       IPv4 uses dot-separated decimal without leading 0's (1.2.3.4)
  *       IPv6 uses colon separated network-order 16 bit hex with leading 0's
@@ -2407,6 +2411,7 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr,
 	case 'M':			/* Colon separated: 00:01:02:03:04:05 */
 	case 'm':			/* Contiguous: 000102030405 */
 					/* [mM]F (FDDI) */
+					/* [mM]U (Uppercase hex) */
 					/* [mM]R (Reverse order; Bluetooth) */
 		return mac_address_string(buf, end, ptr, spec, fmt);
 	case 'I':			/* Formatted IP supported
Signature · 1 line
2.34.1
#8 lore raw

PATCH v4 2/2 usb: gadget: u_ether: Fix host MAC address case

The CDC-ECM specification [1] requires to send the host MAC address as an uppercase hexadecimal string in chapter "5.4 Ethernet Networking Functional Descriptor":
The Unicode character is chosen from the set of values 30h through 39h and 41h through 46h (0-9 and A-F).

However, snprintf(.., "%pm", ..) generates a lowercase MAC address string. While most host drivers are tolerant to this, UsbNcm.sys on Windows 10 is not. Instead it uses a different MAC address with all bytes set to zero including and after the first byte containing a lowercase letter. On Windows 11 Microsoft fixed it, but apparently they did not backport the fix.

This change fixes the issue by using "%pmU" to generate an uppercase hex string to comply with the specification.

[1]: https://www.usb.org/document-library/class-definitions-communication-devices-12, file ECM120.pdf

Fixes bcd4a1c40bee ("usb: gadget: u_ether: construct with default values and add setters/getters")

---

Changes since v3: None

Changes since v2:
* Add uppercase MAC address format string and use that instead of manually uppercasing the resulting MAC address string.

Changes since v1:
* Fixed checkpatch.pl warnings

 drivers/usb/gadget/function/u_ether.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/usb/gadget/function/u_ether.c +1 -1
index 6956ad8ba8dd..70e6b825654c 100644
@@ -963,7 +963,7 @@ int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len)
 		return -EINVAL;
 
 	dev = netdev_priv(net);
-	snprintf(host_addr, len, "%pm", dev->host_mac);
+	snprintf(host_addr, len, "%pmU", dev->host_mac);
 
 	return strlen(host_addr);
 }
Signature · 1 line
2.34.1
#9 lore raw

Re: [PATCH v4 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

In reply to Konrad Gräfe
Quoted 3 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Why not teaching %ph to provide an uppercase? Would be much more useful than this.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#10 lore raw

PATCH v5 1/1 usb: gadget: u_ether: Fix host MAC address case

In reply to Konrad Gräfe

The CDC-ECM specification [1] requires to send the host MAC address as an uppercase hexadecimal string in chapter "5.4 Ethernet Networking Functional Descriptor":
The Unicode character is chosen from the set of values 30h through 39h and 41h through 46h (0-9 and A-F).

However, snprintf(.., "%pm", ..) generates a lowercase MAC address string. While most host drivers are tolerant to this, UsbNcm.sys on Windows 10 is not. Instead it uses a different MAC address with all bytes set to zero including and after the first byte containing a lowercase letter. On Windows 11 Microsoft fixed it, but apparently they did not backport the fix.

This change fixes the issue by upper-casing the MAC to comply with the specification.

[1]: https://www.usb.org/document-library/class-definitions-communication-devices-12, file ECM120.pdf

Fixes bcd4a1c40bee ("usb: gadget: u_ether: construct with default values and add setters/getters")

---
Changes since v4:
* Use string_upper() instead of a special format string

Changes since v3: None

Changes since v2:
* Add uppercase MAC address format string and use that instead of manually uppercasing the resulting MAC address string.

Changes since v1:
* Fixed checkpatch.pl warnings

 drivers/usb/gadget/function/u_ether.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/usb/gadget/function/u_ether.c +3 -0
index 6956ad8ba8dd..a366abb45623 100644
@@ -17,6 +17,7 @@
 #include <linux/etherdevice.h>
 #include <linux/ethtool.h>
 #include <linux/if_vlan.h>
+#include <linux/string_helpers.h>
 #include <linux/usb/composite.h>
 
 #include "u_ether.h"
@@ -965,6 +966,8 @@ int gether_get_host_addr_cdc(struct net_device *net, char *host_addr, int len)
 	dev = netdev_priv(net);
 	snprintf(host_addr, len, "%pm", dev->host_mac);
 
+	string_upper(host_addr, host_addr);
+
 	return strlen(host_addr);
 }
 EXPORT_SYMBOL_GPL(gether_get_host_addr_cdc);
Signature · 1 line
2.34.1
#11 lore raw

Re: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

In reply to Konrad Gräfe
Quoted 29 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

---
Added in v3

 lib/vsprintf.c | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)
lib/vsprintf.c +2 -1
index be71a03c936a..8aee1caabd9e 100644
@@ -1269,9 +1269,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 {
 	char mac_addr[sizeof("xx:xx:xx:xx:xx:xx")];
 	char *p = mac_addr;
-	int i;
+	int i, pos;
 	char separator;
 	bool reversed = false;
+	bool uppercase = false;
 
 	if (check_pointer(&buf, end, addr, spec))
 		return buf;

lib/vsprintf.c char *mac_address_string(char *buf, char *end, u8 *addr, · line 1287

Quoted 6 lines
lib/vsprintf.c +3 -0
@@ -1281,6 +1282,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 		separator = '-';
 		break;
 
+	case 'U':
+		uppercase = true;
+		break;

No documentation update as well?

thanks,

greg k-h

#12 lore raw

Re: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

In reply to Konrad Gräfe
Quoted 5 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Why cc stable?

Quoted 6 lines

---
Added in v3

 lib/vsprintf.c | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

The diffstat here, or for some other patch in the same series, definitely ought to mention lib/test_printf.c.

Quoted 17 lines
lib/vsprintf.c +2 -1
index be71a03c936a..8aee1caabd9e 100644
@@ -1269,9 +1269,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 {
 	char mac_addr[sizeof("xx:xx:xx:xx:xx:xx")];
 	char *p = mac_addr;
-	int i;
+	int i, pos;
 	char separator;
 	bool reversed = false;
+	bool uppercase = false;
 
 	if (check_pointer(&buf, end, addr, spec))
 		return buf;

lib/vsprintf.c char *mac_address_string(char *buf, char *end, u8 *addr, · line 1291

Quoted 10 lines
lib/vsprintf.c +4 -0
@@ -1281,6 +1282,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 		separator = '-';
 		break;
 
+	case 'U':
+		uppercase = true;
+		break;
+
 	case 'R':
 		reversed = true;
 		fallthrough;

This seems broken, and I'm surprised the compiler doesn't warn about separator possibly being uninitialized further down. I'm also surprised your testing hasn't caught this. For reference, the full switch statement is currently

        switch (fmt[1]) {
        case 'F':
                separator = '-';
                break;

        case 'R':
                reversed = true;
                fallthrough;

        default:
                separator = ':';
                break;
        }
Quoted 4 lines
lib/vsprintf.c +0 -0
@@ -1292,9 +1297,14 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 
 	for (i = 0; i < 6; i++) {
 		if (reversed)

lib/vsprintf.c char *mac_address_string(char *buf, char *end, u8 *addr, · line 1307

Quoted 10 lines
lib/vsprintf.c +7 -2
-			p = hex_byte_pack(p, addr[5 - i]);
+			pos = 5 - i;
+		else
+			pos = i;
+
+		if (uppercase)
+			p = hex_byte_pack_upper(p, addr[pos]);
 		else
-			p = hex_byte_pack(p, addr[i]);
+			p = hex_byte_pack(p, addr[pos]);

I think this becomes quite hard to follow. We have string_upper() in linux/string_helpers.h, so I'd rather just leave this loop alone and do

  if (uppercase)
    string_upper(mac_addr, mac_addr);

after the nul-termination.

Rasmus

#13 lore raw

Re: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

Quoted 8 lines · Rasmus Villemoes
Quoted 5 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Why cc stable?

I believe the second patch matches the criteria but it uses this one.

Quoted 83 lines
Quoted 7 lines

---
Added in v3

  lib/vsprintf.c | 18 +++++++++++++++---
  1 file changed, 15 insertions(+), 3 deletions(-)

The diffstat here, or for some other patch in the same series, definitely ought to mention lib/test_printf.c.

Quoted 17 lines
lib/vsprintf.c +2 -1
index be71a03c936a..8aee1caabd9e 100644
@@ -1269,9 +1269,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
  {
  	char mac_addr[sizeof("xx:xx:xx:xx:xx:xx")];
  	char *p = mac_addr;
-	int i;
+	int i, pos;
  	char separator;
  	bool reversed = false;
+	bool uppercase = false;
 
  	if (check_pointer(&buf, end, addr, spec))
  		return buf;

lib/vsprintf.c char *mac_address_string(char *buf, char *end, u8 *addr, · line 1291

Quoted 10 lines
lib/vsprintf.c +4 -0
@@ -1281,6 +1282,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
  		separator = '-';
  		break;
 
+	case 'U':
+		uppercase = true;
+		break;
+
  	case 'R':
  		reversed = true;
  		fallthrough;

This seems broken, and I'm surprised the compiler doesn't warn about separator possibly being uninitialized further down. I'm also surprised your testing hasn't caught this. For reference, the full switch statement is currently

         switch (fmt[1]) {
         case 'F':
                 separator = '-';
                 break;

         case 'R':
                 reversed = true;
                 fallthrough;

         default:
                 separator = ':';
                 break;
         }
Quoted 4 lines
lib/vsprintf.c +0 -0
@@ -1292,9 +1297,14 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 
  	for (i = 0; i < 6; i++) {
  		if (reversed)

lib/vsprintf.c char *mac_address_string(char *buf, char *end, u8 *addr, · line 1307

Quoted 10 lines
lib/vsprintf.c +7 -2
-			p = hex_byte_pack(p, addr[5 - i]);
+			pos = 5 - i;
+		else
+			pos = i;
+
+		if (uppercase)
+			p = hex_byte_pack_upper(p, addr[pos]);
  		else
-			p = hex_byte_pack(p, addr[i]);
+			p = hex_byte_pack(p, addr[pos]);

I think this becomes quite hard to follow. We have string_upper() in linux/string_helpers.h, so I'd rather just leave this loop alone and do

   if (uppercase)
     string_upper(mac_addr, mac_addr);

after the nul-termination.

Rasmus

#14 lore raw

Re: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

In reply to Rasmus Villemoes

Hello Rasmus, Konrad, *,

Quoted 39 lines · Rasmus Villemoes
Quoted 5 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Why cc stable?

Quoted 6 lines

---
Added in v3

 lib/vsprintf.c | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)  

The diffstat here, or for some other patch in the same series, definitely ought to mention lib/test_printf.c.

Quoted 20 lines
lib/vsprintf.c +2 -1
index be71a03c936a..8aee1caabd9e 100644
@@ -1269,9 +1269,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 {
 	char mac_addr[sizeof("xx:xx:xx:xx:xx:xx")];
 	char *p = mac_addr;
-	int i;
+	int i, pos;
 	char separator;
 	bool reversed = false;
+	bool uppercase = false;
 
 	if (check_pointer(&buf, end, addr, spec))
 		return buf;
@@ -1281,6 +1282,10 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 		separator = '-';
 		break;
 
Quoted 12 lines

lib/vsprintf.c char *mac_address_string(char *buf, char *end, u8 *addr, · line 1291

Quoted 7 lines
lib/vsprintf.c +4 -0
+	case 'U':
+		uppercase = true;
+		break;
+
 	case 'R':
 		reversed = true;
 		fallthrough;  

This seems broken, and I'm surprised the compiler doesn't warn about separator possibly being uninitialized further down. I'm also surprised your testing hasn't caught this. For reference, the full switch statement is currently

Compiler (gcc) does not warn because of Makefile:

1038 # Enabled with W=2, disabled by default as noisy 1039 ifdef CONFIG_CC_IS_GCC
1040 KBUILD_CFLAGS += -Wno-maybe-uninitialized 1041 endif

With this commented:

lib/vsprintf.c: In function ‘mac_address_string’:
lib/vsprintf.c:1310:30: warning: ‘separator’ may be used uninitialized [-Wmaybe-uninitialized]

   1310 |                         *p++ = separator;
        |                         ~~~~~^~~~~~~~~~~
  lib/vsprintf.c:1273:14: note: ‘separator’ was declared here
   1273 |         char separator;
        |              ^~~~~~~~~

Regards,
Peter

Quoted 40 lines
        switch (fmt[1]) {
        case 'F':
                separator = '-';
                break;

        case 'R':
                reversed = true;
                fallthrough;

        default:
                separator = ':';
                break;
        }
Quoted 4 lines
lib/vsprintf.c +0 -0
@@ -1292,9 +1297,14 @@ char *mac_address_string(char *buf, char *end, u8 *addr,
 
 	for (i = 0; i < 6; i++) {
 		if (reversed)

lib/vsprintf.c char *mac_address_string(char *buf, char *end, u8 *addr, · line 1307

Quoted 10 lines
lib/vsprintf.c +7 -2
-			p = hex_byte_pack(p, addr[5 - i]);
+			pos = 5 - i;
+		else
+			pos = i;
+
+		if (uppercase)
+			p = hex_byte_pack_upper(p, addr[pos]);
 		else
-			p = hex_byte_pack(p, addr[i]);
+			p = hex_byte_pack(p, addr[pos]);  

I think this becomes quite hard to follow. We have string_upper() in linux/string_helpers.h, so I'd rather just leave this loop alone and do

  if (uppercase)
    string_upper(mac_addr, mac_addr);

after the nul-termination.

Rasmus

#15 lore raw

Re: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

In reply to Konrad Gräfe
Quoted 3 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Thinking more about it, I'm not sure this is appropriate, not for a single user like this. vsprintf() should not and cannot satisfy all possible string formatting requirements for the whole kernel. The %pX extensions are convenient for use with printk() and friends where one needs what in other languages would be "string interpolation" (because then the caller doesn't need to deal with temporary stack buffers and pass them as %s arguments), but for single items like this, snprintf() is not necessarily the right tool for the job.

In this case, the caller can just as well call string_upper() on the result, or not use sprintf() at all and do a tiny loop with hex_byte_pack_upper().

Rasmus

#16 lore raw

Re: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

Quoted 5 lines · Rasmus Villemoes
Quoted 3 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Quoted 8 lines

Thinking more about it, I'm not sure this is appropriate, not for a single user like this. vsprintf() should not and cannot satisfy all possible string formatting requirements for the whole kernel. The %pX extensions are convenient for use with printk() and friends where one needs what in other languages would be "string interpolation" (because then the caller doesn't need to deal with temporary stack buffers and pass them as %s arguments), but for single items like this, snprintf() is not necessarily the right tool for the job.

But sprintf() already creates mac address strings today, adding yet-another-modifier makes it so that we don't have to hand-roll this type of logic in the individual drivers that require it.

thanks,

greg k-h

#17 lore raw

RE: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

In reply to Rasmus Villemoes

From: Rasmus Villemoes

Quoted 16 lines

Sent: 28 April 2023 07:57

Quoted 3 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Thinking more about it, I'm not sure this is appropriate, not for a single user like this. vsprintf() should not and cannot satisfy all possible string formatting requirements for the whole kernel. The %pX extensions are convenient for use with printk() and friends where one needs what in other languages would be "string interpolation" (because then the caller doesn't need to deal with temporary stack buffers and pass them as %s arguments), but for single items like this, snprintf() is not necessarily the right tool for the job.

Quoted 3 lines

In this case, the caller can just as well call string_upper() on the result, or not use sprintf() at all and do a tiny loop with hex_byte_pack_upper().

Or snprintf with "%02X:%02X:%02X:%02X:%02X:%02X".

David

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

#18 lore raw

Re: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

Quoted 16 lines · David Laight

From: Rasmus Villemoes

Quoted 15 lines

Sent: 28 April 2023 07:57

Quoted 3 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Thinking more about it, I'm not sure this is appropriate, not for a single user like this. vsprintf() should not and cannot satisfy all possible string formatting requirements for the whole kernel. The %pX extensions are convenient for use with printk() and friends where one needs what in other languages would be "string interpolation" (because then the caller doesn't need to deal with temporary stack buffers and pass them as %s arguments), but for single items like this, snprintf() is not necessarily the right tool for the job.

Quoted 5 lines
Quoted 3 lines

In this case, the caller can just as well call string_upper() on the result, or not use sprintf() at all and do a tiny loop with hex_byte_pack_upper().

Or snprintf with "%02X:%02X:%02X:%02X:%02X:%02X".

Of course this is a step back. Why? Have you read actually what we have in %p extensions already?

Also, what about stack?

Entire %pm/M exists due to reversed order. Otherwise it's an alias to %6phD or alike.

Signature · 2 lines
With Best Regards,
Andy Shevchenko
maintainer
#19 lore raw

Re: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

In reply to Rasmus Villemoes
Quoted 14 lines · Rasmus Villemoes
Quoted 3 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Thinking more about it, I'm not sure this is appropriate, not for a single user like this. vsprintf() should not and cannot satisfy all possible string formatting requirements for the whole kernel. The %pX extensions are convenient for use with printk() and friends where one needs what in other languages would be "string interpolation" (because then the caller doesn't need to deal with temporary stack buffers and pass them as %s arguments), but for single items like this, snprintf() is not necessarily the right tool for the job.

Quoted 2 lines

In this case, the caller can just as well call string_upper() on the result

I tend to agree with Rasmus. string_upper() is a super-easy solution. One user does not look worth adding all the churn into vsprintf().

Best Regards,
Petr

#20 lore raw

Re: [PATCH v3 1/2] vsprintf: Add %p[mM]U for uppercase MAC address

Quoted 15 lines · Petr Mladek
Quoted 14 lines · Rasmus Villemoes
Quoted 3 lines · Konrad Gräfe

The CDC-ECM specification requires an USB gadget to send the host MAC address as uppercase hex string. This change adds the appropriate modifier.

Thinking more about it, I'm not sure this is appropriate, not for a single user like this. vsprintf() should not and cannot satisfy all possible string formatting requirements for the whole kernel. The %pX extensions are convenient for use with printk() and friends where one needs what in other languages would be "string interpolation" (because then the caller doesn't need to deal with temporary stack buffers and pass them as %s arguments), but for single items like this, snprintf() is not necessarily the right tool for the job.

Quoted 8 lines
Quoted 2 lines

In this case, the caller can just as well call string_upper() on the result

I tend to agree with Rasmus. string_upper() is a super-easy solution. One user does not look worth adding all the churn into vsprintf().

Best Regards,
Petr

I do agree as well. That would basically be v1 [1] without the hand-crafted string_upper(). (I didn't know the function.)

Regards,
Konrad

[1]: https://lore.kernel.org/linux-usb/[email protected]/