From: Michael Neuling <mikey@neuling.org>
To: benh@kernel.crashing.org
Cc: Stephen Rothwell <sfr@canb.auug.org.au>,
linuxppc-dev@lists.ozlabs.org, Anton Blanchard <anton@samba.org>
Subject: [PATCH v3 3/3] powerpc: Correctly context switch DSCR on POWER8
Date: Fri, 09 Aug 2013 15:49:20 +1000 [thread overview]
Message-ID: <20158.1376027360@ale.ozlabs.ibm.com> (raw)
In-Reply-To: <30983.1375700435@ale.ozlabs.ibm.com>
POWER8 allows the DSCR to be accessed directly from userspace via a new SPR
number 0x3 (Rather than 0x11. DSCR SPR number 0x11 is still used on POWER8 but
like POWER7, is only accessible in HV and OS modes). Currently, we allow this
by setting H/FSCR DSCR bit on boot.
Unfortunately this doesn't work, as the kernel needs to see the DSCR change so
that it knows to no longer restore the system wide version of DSCR on context
switch (ie. to set thread.dscr_inherit).
This clears the H/FSCR DSCR bit initially. If a process then accesses the DSCR
(via SPR 0x3), it'll trap into the kernel where we set thread.dscr_inherit in
facility_unavailable_exception().
We also change _switch() so that we set or clear the H/FSCR DSCR bit based on
the thread.dscr_inherit.
Signed-off-by: Michael Neuling <mikey@neuling.org>
---
v3: remove unnecessary clearing of H/FSCR at boot.
diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
index ab15b8d..4674fe6 100644
--- a/arch/powerpc/kernel/entry_64.S
+++ b/arch/powerpc/kernel/entry_64.S
@@ -584,9 +584,34 @@ BEGIN_FTR_SECTION
ld r7,DSCR_DEFAULT@toc(2)
ld r0,THREAD_DSCR(r4)
cmpwi r6,0
+ li r8, FSCR_DSCR
bne 1f
ld r0,0(r7)
-1: cmpd r0,r25
+ b 3f
+1:
+ BEGIN_FTR_SECTION_NESTED(70)
+ mfspr r6, SPRN_FSCR
+ or r6, r6, r8
+ mtspr SPRN_FSCR, r6
+ BEGIN_FTR_SECTION_NESTED(69)
+ mfspr r6, SPRN_HFSCR
+ or r6, r6, r8
+ mtspr SPRN_HFSCR, r6
+ END_FTR_SECTION_NESTED(CPU_FTR_HVMODE, CPU_FTR_HVMODE, 69)
+ b 4f
+ END_FTR_SECTION_NESTED(CPU_FTR_ARCH_207S, CPU_FTR_ARCH_207S, 70)
+3:
+ BEGIN_FTR_SECTION_NESTED(70)
+ mfspr r6, SPRN_FSCR
+ andc r6, r6, r8
+ mtspr SPRN_FSCR, r6
+ BEGIN_FTR_SECTION_NESTED(69)
+ mfspr r6, SPRN_HFSCR
+ andc r6, r6, r8
+ mtspr SPRN_HFSCR, r6
+ END_FTR_SECTION_NESTED(CPU_FTR_HVMODE, CPU_FTR_HVMODE, 69)
+ END_FTR_SECTION_NESTED(CPU_FTR_ARCH_207S, CPU_FTR_ARCH_207S, 70)
+4: cmpd r0,r25
beq 2f
mtspr SPRN_DSCR,r0
2:
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index bf33c22..e435bc0 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -44,9 +44,7 @@
#include <asm/machdep.h>
#include <asm/rtas.h>
#include <asm/pmc.h>
-#ifdef CONFIG_PPC32
#include <asm/reg.h>
-#endif
#ifdef CONFIG_PMAC_BACKLIGHT
#include <asm/backlight.h>
#endif
@@ -1296,43 +1294,54 @@ void vsx_unavailable_exception(struct pt_regs *regs)
die("Unrecoverable VSX Unavailable Exception", regs, SIGABRT);
}
+#ifdef CONFIG_PPC64
void facility_unavailable_exception(struct pt_regs *regs)
{
static char *facility_strings[] = {
- "FPU",
- "VMX/VSX",
- "DSCR",
- "PMU SPRs",
- "BHRB",
- "TM",
- "AT",
- "EBB",
- "TAR",
+ [FSCR_FP_LG] = "FPU",
+ [FSCR_VECVSX_LG] = "VMX/VSX",
+ [FSCR_DSCR_LG] = "DSCR",
+ [FSCR_PM_LG] = "PMU SPRs",
+ [FSCR_BHRB_LG] = "BHRB",
+ [FSCR_TM_LG] = "TM",
+ [FSCR_EBB_LG] = "EBB",
+ [FSCR_TAR_LG] = "TAR",
};
- char *facility, *prefix;
+ char *facility = "unknown";
u64 value;
+ u8 status;
+ bool hv;
- if (regs->trap == 0xf60) {
- value = mfspr(SPRN_FSCR);
- prefix = "";
- } else {
+ hv = (regs->trap == 0xf80);
+ if (hv)
value = mfspr(SPRN_HFSCR);
- prefix = "Hypervisor ";
+ else
+ value = mfspr(SPRN_FSCR);
+
+ status = value >> 56;
+ if (status == FSCR_DSCR_LG) {
+ /* User is acessing the DSCR. Set the inherit bit and allow
+ * the user to set it directly in future by setting via the
+ * H/FSCR DSCR bit.
+ */
+ current->thread.dscr_inherit = 1;
+ if (hv)
+ mtspr(SPRN_HFSCR, value | HFSCR_DSCR);
+ else
+ mtspr(SPRN_FSCR, value | FSCR_DSCR);
+ return;
}
- value = value >> 56;
+ if ((status < ARRAY_SIZE(facility_strings)) &&
+ facility_strings[status])
+ facility = facility_strings[status];
/* We restore the interrupt state now */
if (!arch_irq_disabled_regs(regs))
local_irq_enable();
- if (value < ARRAY_SIZE(facility_strings))
- facility = facility_strings[value];
- else
- facility = "unknown";
-
pr_err("%sFacility '%s' unavailable, exception at 0x%lx, MSR=%lx\n",
- prefix, facility, regs->nip, regs->msr);
+ hv ? "Hypervisor " : "", facility, regs->nip, regs->msr);
if (user_mode(regs)) {
_exception(SIGILL, regs, ILL_ILLOPC, regs->nip);
@@ -1341,6 +1350,7 @@ void facility_unavailable_exception(struct pt_regs *regs)
die("Unexpected facility unavailable exception", regs, SIGABRT);
}
+#endif
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
next prev parent reply other threads:[~2013-08-09 5:49 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-08-05 7:28 [PATCH 1/3] powerpc: Rework setting up H/FSCR bit definitions Michael Neuling
2013-08-05 7:28 ` [PATCH 2/3] powerpc: Fix hypervisor facility unavaliable vector number Michael Neuling
2013-08-05 7:28 ` [PATCH 3/3] powerpc: Correctly context switch DSCR on POWER8 Michael Neuling
2013-08-05 8:42 ` Stephen Rothwell
2013-08-05 9:26 ` Michael Neuling
2013-08-05 10:16 ` Michael Neuling
2013-08-05 11:00 ` [PATCH v2 " Michael Neuling
2013-08-09 5:49 ` Michael Neuling [this message]
2013-08-09 3:46 ` [PATCH v2 1/3] powerpc: Rework setting up H/FSCR bit definitions Michael Neuling
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20158.1376027360@ale.ozlabs.ibm.com \
--to=mikey@neuling.org \
--cc=anton@samba.org \
--cc=benh@kernel.crashing.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=sfr@canb.auug.org.au \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).