Date: Tue, 16 Jan 2018 01:42:18 +0200 From: Konstantin Belousov <kostikbel@gmail.com> To: Nathan Whitehorn <nwhitehorn@freebsd.org> Cc: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: Re: svn commit: r327950 - in head/sys/powerpc: aim include powerpc ps3 Message-ID: <20180115234218.GN1684@kib.kiev.ua> In-Reply-To: <bb27ba01-8383-6b85-8b2b-65227ff46efc@freebsd.org> References: <20180114175211.GD1684@kib.kiev.ua> <b2b1bf30-177b-af30-54ce-f484224bb2ad@freebsd.org> <f4b44b69-7b06-6b5a-c17c-31bd46ca1af0@freebsd.org> <e04bc7a6-fa77-9ca0-2aff-dc29c543c9a1@freebsd.org> <20180115111812.GF1684@kib.kiev.ua> <f6350c61-55d1-9bf7-c4b3-e10fb329a42a@freebsd.org> <20180115170603.GJ1684@kib.kiev.ua> <9e5554d7-6a0c-5910-8cb6-74f98259536f@freebsd.org> <20180115175335.GK1684@kib.kiev.ua> <bb27ba01-8383-6b85-8b2b-65227ff46efc@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Jan 15, 2018 at 03:20:49PM -0800, Nathan Whitehorn wrote:
> Fair enough. Here's a patch with a new flag (DIRECT_MAP_AVAILABLE). I've
> also retooled the sfbuf code to use this rather than its own flags that
> mean the same things. The sparc64 part of the patch is untested.
> -Nathan
> Index: amd64/include/vmparam.h
> ===================================================================
> --- amd64/include/vmparam.h (revision 328006)
> +++ amd64/include/vmparam.h (working copy)
> @@ -190,6 +190,7 @@
> * because the result is not actually accessed until later, but the early
> * vt fb startup needs to be reworked.
> */
> +#define DIRECT_MAP_AVAILABLE 1
> #define PHYS_TO_DMAP(x) ({ \
> KASSERT(dmaplimit == 0 || (x) < dmaplimit, \
> ("physical address %#jx not covered by the DMAP", \
> Index: arm64/include/vmparam.h
> ===================================================================
> --- arm64/include/vmparam.h (revision 328006)
> +++ arm64/include/vmparam.h (working copy)
> @@ -176,6 +176,7 @@
> #define VIRT_IN_DMAP(va) ((va) >= DMAP_MIN_ADDRESS && \
> (va) < (dmap_max_addr))
>
> +#define DIRECT_MAP_AVAILABLE
Just define, or define it to 1 ?
> #define PHYS_TO_DMAP(pa) \
> ({ \
> KASSERT(PHYS_IN_DMAP(pa), \
> Index: dev/efidev/efirt.c
> ===================================================================
> --- dev/efidev/efirt.c (revision 328006)
> +++ dev/efidev/efirt.c (working copy)
> @@ -115,6 +115,11 @@
> return (0);
> }
> efi_systbl = (struct efi_systbl *)PHYS_TO_DMAP(efi_systbl_phys);
> + if (efi_systbl == NULL) {
> + if (bootverbose)
> + printf("EFI systbl not mapped in kernel VA\n");
> + return (0);
> + }
Is this chunk still needed ?
> if (efi_systbl->st_hdr.th_sig != EFI_SYSTBL_SIG) {
> efi_systbl = NULL;
> if (bootverbose)
> Index: kern/subr_sfbuf.c
> ===================================================================
> --- kern/subr_sfbuf.c (revision 328006)
> +++ kern/subr_sfbuf.c (working copy)
> @@ -88,8 +88,8 @@
> vm_offset_t sf_base;
> int i;
>
> -#ifdef SFBUF_OPTIONAL_DIRECT_MAP
> - if (SFBUF_OPTIONAL_DIRECT_MAP)
> +#ifdef DIRECT_MAP_AVAILABLE
> + if (DIRECT_MAP_AVAILABLE)
> return;
Would it make sense to define the symbol on all other arches as 0 then,
and remove #ifdef ? Returning to your initial proposal of relying on the
compiler optimiing if (0) block; out.
Also, just curious, why did you spelled DMAP as DIRECT_MAP ?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20180115234218.GN1684>