[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 joe.harvell at netscout dot com changed: What|Removed |Added CC||joe.harvell at netscout

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 --- Comment #10 from stephane.goujet at wanadoo dot fr --- (In reply to stephane.goujet from comment #9) > 2. There are inconsistencies in the Warning: Another inconsistency: 2.c The documentation of the packed attribute says "This attribute,

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 stephane.goujet at wanadoo dot fr changed: What|Removed |Added Status|RESOLVED|UNCONFIRMED

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 --- Comment #8 from Frederic Marchal --- Regarding the pointer vs pointeur typo in French, it will be fixed with the next translation update. Thanks for reporting it.

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|---

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 --- Comment #6 from stephane.goujet at wanadoo dot fr --- (In reply to Andreas Schwab from comment #5) > The packed attribute forces the alignment to 1, so there is no requirement > for its address to be aligned for its type. So one could say

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 --- Comment #5 from Andreas Schwab --- The packed attribute forces the alignment to 1, so there is no requirement for its address to be aligned for its type.

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 Eric Gallager changed: What|Removed |Added CC||egallager at gcc dot gnu.org,

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 --- Comment #3 from stephane.goujet at wanadoo dot fr --- (In reply to Mikael Pettersson from comment #2) > IMO the warning is correct. Simply _creating_ a misaligned pointer is > undefined behaviour, and there are machines where it cannot be

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 --- Comment #2 from Mikael Pettersson --- (In reply to stephane.goujet from comment #0) > test2.c: In function 'main': > test2.c:9:5: warning: converting a packed 'struct S' pointer (alignment 1) > to a 'int' pointer (alignment 4) may result in

[Bug c/93910] -Waddress-of-packed-member triggered without actual access to a member or the address of a member

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93910 --- Comment #1 from Andreas Schwab --- You are still converting an unaligned pointer to an aligned pointer, and the address of a struct is the same as the address if its first member (with suitable conversion).