On Sun, Sep 13, 2020 at 1:46 PM Arnd Bergmann <arnd@xxxxxxxx> wrote: > > On Sun, Sep 13, 2020 at 8:50 AM Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote: > > > > On Sat, Sep 12, 2020 at 02:49:05PM +0200, Arnd Bergmann wrote: > > > fs/quota/compat.c: dqblk = compat_alloc_user_space(sizeof(struct if_dqblk)); > > > fs/quota/compat.c: dqblk = compat_alloc_user_space(sizeof(struct if_dqblk)); > > > fs/quota/compat.c: fsqstat = compat_alloc_user_space(sizeof(struct > > > fs_quota_stat)); > > > > I sent this out a while ago, an Al has it in a branch, but not in > > linux-next: > > > > https://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs.git/log/?h=work.quota-compat > > Nice! Aside from already being queued, your patch is also nicer than > my version, and it makes it trivial to fix it for arm oabi as well by adding > > #ifdef CONFIG_OABI_COMPAT > #define compat_need_64bit_alignment_fixup in_oabi_syscall > #endif > > to arch/arm/include/asm/compat.h > > I had considered fixing that case for arch/arm as well but it ended up being > harder to do in my version. Unfortunately, the commit b902bfb3f0e "arm64: stop using <asm/compat.h> directly" seems to introduce a circular header file inclusion between linux/compat.h and asm/stat.h, breaking arm64 compilation. Moving the compat_u64/compat_s64 definitions to include/asm-generic/compat.h works fine though. Arnd
