Patchwork sparc64: fix unused-but-set-variable warnings

login
register
mail settings
Submitter David Miller
Date Feb. 27, 2011, 7:41 a.m.
Message ID <20110226.234112.48497619.davem@davemloft.net>
Download mbox | patch
Permalink /patch/84678/
State Accepted
Delegated to: David Miller
Headers show

Comments

David Miller - Feb. 27, 2011, 7:41 a.m.
Here is the patch I will push upstream, it passes sparc64_defconfig,
sparc64_defconfig + CONFIG_SMP=n, allnoconfig, and allmodconfig builds.

--------------------
sparc64: Fix build errors with gcc-4.6.0

Most of the warnings emitted (we fail arch/sparc file
builds with -Werror) were legitimate but harmless, however
one case (n2_pcr_write) was a genuine bug.

Based almost entirely upon a patch by Sam Ravnborg.

Reported-by: Dennis Gilmore <dennis@ausil.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc/kernel/iommu.c      |    3 ---
 arch/sparc/kernel/ldc.c        |   28 ++++++++++++++++++----------
 arch/sparc/kernel/pci.c        |    1 +
 arch/sparc/kernel/pci_common.c |   11 +++++++----
 arch/sparc/kernel/pci_fire.c   |    2 --
 arch/sparc/kernel/pci_schizo.c |    4 +---
 arch/sparc/kernel/pci_sun4v.c  |    3 +--
 arch/sparc/kernel/pcr.c        |    2 +-
 arch/sparc/kernel/ptrace_64.c  |    3 ++-
 arch/sparc/kernel/smp_64.c     |   11 ++++-------
 arch/sparc/kernel/traps_64.c   |    3 +--
 11 files changed, 36 insertions(+), 35 deletions(-)
Sam Ravnborg - Feb. 27, 2011, 1:12 p.m.
On Sat, Feb 26, 2011 at 11:41:12PM -0800, David Miller wrote:
> 
> Here is the patch I will push upstream, it passes sparc64_defconfig,
> sparc64_defconfig + CONFIG_SMP=n, allnoconfig, and allmodconfig builds.

Looks good.

> +	/* We should probably check for hypervisor errors here and
> +	 * reset the LDC channel if we get one.
> +	 */

Coding style purist would say that the comment should look like this:

> +     /*
> +	 * We should probably check for hypervisor errors here and
> +      * reset the LDC channel if we get one.
> +      */

I do this for new stuff - and as part of other changes when I do
general clean-ups. It does not warrant an independet patch.

Patch is already pushed to kernel.org - so I do not expect any update.

	Sam
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
David Miller - Feb. 27, 2011, 11:24 p.m.
From: Sam Ravnborg <sam@ravnborg.org>
Date: Sun, 27 Feb 2011 14:12:19 +0100

> On Sat, Feb 26, 2011 at 11:41:12PM -0800, David Miller wrote:
>> +	/* We should probably check for hypervisor errors here and
>> +	 * reset the LDC channel if we get one.
>> +	 */
> 
> Coding style purist would say that the comment should look like this:

Not in my coding style.  That initial comment line is wasted space.

And a coding style purist would understand why that very rule
goes aginst the proported reasoning behind the coding style
rules (eliminating wasted space so more real information fits on
the screen at once) and would thus have the rules fixed.

:-)

I will reject any future changes from you which add wasted vertical
space like this, so please don't get into the habit of trying to inject
this folly into either sparc or the networking, thanks.
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Sam Ravnborg - Feb. 28, 2011, 5:49 a.m.
On Sun, Feb 27, 2011 at 03:24:33PM -0800, David Miller wrote:
> From: Sam Ravnborg <sam@ravnborg.org>
> Date: Sun, 27 Feb 2011 14:12:19 +0100
> 
> > On Sat, Feb 26, 2011 at 11:41:12PM -0800, David Miller wrote:
> >> +	/* We should probably check for hypervisor errors here and
> >> +	 * reset the LDC channel if we get one.
> >> +	 */
> > 
> > Coding style purist would say that the comment should look like this:
> 
> Not in my coding style.  That initial comment line is wasted space.
> 
> And a coding style purist would understand why that very rule
> goes aginst the proported reasoning behind the coding style
> rules (eliminating wasted space so more real information fits on
> the screen at once) and would thus have the rules fixed.
> 
> :-)
> 
> I will reject any future changes from you which add wasted vertical
> space like this, so please don't get into the habit of trying to inject
> this folly into either sparc or the networking, thanks.

OK

(And one day I may even start to use "//" for comments :-) )

	Sam
--
To unsubscribe from this list: send the line "unsubscribe sparclinux" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/arch/sparc/kernel/iommu.c b/arch/sparc/kernel/iommu.c
index 72509d0..6f01e8c 100644
--- a/arch/sparc/kernel/iommu.c
+++ b/arch/sparc/kernel/iommu.c
@@ -333,13 +333,10 @@  static void dma_4u_free_coherent(struct device *dev, size_t size,
 				 void *cpu, dma_addr_t dvma)
 {
 	struct iommu *iommu;
-	iopte_t *iopte;
 	unsigned long flags, order, npages;
 
 	npages = IO_PAGE_ALIGN(size) >> IO_PAGE_SHIFT;
 	iommu = dev->archdata.iommu;
-	iopte = iommu->page_table +
-		((dvma - iommu->page_table_map_base) >> IO_PAGE_SHIFT);
 
 	spin_lock_irqsave(&iommu->lock, flags);
 
diff --git a/arch/sparc/kernel/ldc.c b/arch/sparc/kernel/ldc.c
index df39a0f..732b0bc 100644
--- a/arch/sparc/kernel/ldc.c
+++ b/arch/sparc/kernel/ldc.c
@@ -790,16 +790,20 @@  static void send_events(struct ldc_channel *lp, unsigned int event_mask)
 static irqreturn_t ldc_rx(int irq, void *dev_id)
 {
 	struct ldc_channel *lp = dev_id;
-	unsigned long orig_state, hv_err, flags;
+	unsigned long orig_state, flags;
 	unsigned int event_mask;
 
 	spin_lock_irqsave(&lp->lock, flags);
 
 	orig_state = lp->chan_state;
-	hv_err = sun4v_ldc_rx_get_state(lp->id,
-					&lp->rx_head,
-					&lp->rx_tail,
-					&lp->chan_state);
+
+	/* We should probably check for hypervisor errors here and
+	 * reset the LDC channel if we get one.
+	 */
+	sun4v_ldc_rx_get_state(lp->id,
+			       &lp->rx_head,
+			       &lp->rx_tail,
+			       &lp->chan_state);
 
 	ldcdbg(RX, "RX state[0x%02lx:0x%02lx] head[0x%04lx] tail[0x%04lx]\n",
 	       orig_state, lp->chan_state, lp->rx_head, lp->rx_tail);
@@ -904,16 +908,20 @@  out:
 static irqreturn_t ldc_tx(int irq, void *dev_id)
 {
 	struct ldc_channel *lp = dev_id;
-	unsigned long flags, hv_err, orig_state;
+	unsigned long flags, orig_state;
 	unsigned int event_mask = 0;
 
 	spin_lock_irqsave(&lp->lock, flags);
 
 	orig_state = lp->chan_state;
-	hv_err = sun4v_ldc_tx_get_state(lp->id,
-					&lp->tx_head,
-					&lp->tx_tail,
-					&lp->chan_state);
+
+	/* We should probably check for hypervisor errors here and
+	 * reset the LDC channel if we get one.
+	 */
+	sun4v_ldc_tx_get_state(lp->id,
+			       &lp->tx_head,
+			       &lp->tx_tail,
+			       &lp->chan_state);
 
 	ldcdbg(TX, " TX state[0x%02lx:0x%02lx] head[0x%04lx] tail[0x%04lx]\n",
 	       orig_state, lp->chan_state, lp->tx_head, lp->tx_tail);
diff --git a/arch/sparc/kernel/pci.c b/arch/sparc/kernel/pci.c
index 4137579..f255382 100644
--- a/arch/sparc/kernel/pci.c
+++ b/arch/sparc/kernel/pci.c
@@ -675,6 +675,7 @@  static void __devinit pci_bus_register_of_sysfs(struct pci_bus *bus)
 		 * humanoid.
 		 */
 		err = sysfs_create_file(&dev->dev.kobj, &dev_attr_obppath.attr);
+		(void) err;
 	}
 	list_for_each_entry(child_bus, &bus->children, node)
 		pci_bus_register_of_sysfs(child_bus);
diff --git a/arch/sparc/kernel/pci_common.c b/arch/sparc/kernel/pci_common.c
index 6c7a33a..6e3874b 100644
--- a/arch/sparc/kernel/pci_common.c
+++ b/arch/sparc/kernel/pci_common.c
@@ -295,14 +295,17 @@  static int sun4v_write_pci_cfg(struct pci_bus *bus_dev, unsigned int devfn,
 	unsigned int bus = bus_dev->number;
 	unsigned int device = PCI_SLOT(devfn);
 	unsigned int func = PCI_FUNC(devfn);
-	unsigned long ret;
 
 	if (config_out_of_range(pbm, bus, devfn, where)) {
 		/* Do nothing. */
 	} else {
-		ret = pci_sun4v_config_put(devhandle,
-				HV_PCI_DEVICE_BUILD(bus, device, func),
-				where, size, value);
+		/* We don't check for hypervisor errors here, but perhaps
+		 * we should and influence our return value depending upon
+		 * what kind of error is thrown.
+		 */
+		pci_sun4v_config_put(devhandle,
+				     HV_PCI_DEVICE_BUILD(bus, device, func),
+				     where, size, value);
 	}
 	return PCIBIOS_SUCCESSFUL;
 }
diff --git a/arch/sparc/kernel/pci_fire.c b/arch/sparc/kernel/pci_fire.c
index efb896d..75dfeb6 100644
--- a/arch/sparc/kernel/pci_fire.c
+++ b/arch/sparc/kernel/pci_fire.c
@@ -214,11 +214,9 @@  static int pci_fire_msi_setup(struct pci_pbm_info *pbm, unsigned long msiqid,
 
 static int pci_fire_msi_teardown(struct pci_pbm_info *pbm, unsigned long msi)
 {
-	unsigned long msiqid;
 	u64 val;
 
 	val = upa_readq(pbm->pbm_regs + MSI_MAP(msi));
-	msiqid = (val & MSI_MAP_EQNUM);
 
 	val &= ~MSI_MAP_VALID;
 
diff --git a/arch/sparc/kernel/pci_schizo.c b/arch/sparc/kernel/pci_schizo.c
index 445a47a..4620eb7 100644
--- a/arch/sparc/kernel/pci_schizo.c
+++ b/arch/sparc/kernel/pci_schizo.c
@@ -1313,7 +1313,7 @@  static int __devinit schizo_pbm_init(struct pci_pbm_info *pbm,
 	const struct linux_prom64_registers *regs;
 	struct device_node *dp = op->dev.of_node;
 	const char *chipset_name;
-	int is_pbm_a, err;
+	int err;
 
 	switch (chip_type) {
 	case PBM_CHIP_TYPE_TOMATILLO:
@@ -1343,8 +1343,6 @@  static int __devinit schizo_pbm_init(struct pci_pbm_info *pbm,
 	 */
 	regs = of_get_property(dp, "reg", NULL);
 
-	is_pbm_a = ((regs[0].phys_addr & 0x00700000) == 0x00600000);
-
 	pbm->next = pci_pbm_root;
 	pci_pbm_root = pbm;
 
diff --git a/arch/sparc/kernel/pci_sun4v.c b/arch/sparc/kernel/pci_sun4v.c
index 743344a..859abfd 100644
--- a/arch/sparc/kernel/pci_sun4v.c
+++ b/arch/sparc/kernel/pci_sun4v.c
@@ -580,7 +580,7 @@  static int __devinit pci_sun4v_iommu_init(struct pci_pbm_info *pbm)
 {
 	static const u32 vdma_default[] = { 0x80000000, 0x80000000 };
 	struct iommu *iommu = pbm->iommu;
-	unsigned long num_tsb_entries, sz, tsbsize;
+	unsigned long num_tsb_entries, sz;
 	u32 dma_mask, dma_offset;
 	const u32 *vdma;
 
@@ -596,7 +596,6 @@  static int __devinit pci_sun4v_iommu_init(struct pci_pbm_info *pbm)
 
 	dma_mask = (roundup_pow_of_two(vdma[1]) - 1UL);
 	num_tsb_entries = vdma[1] / IO_PAGE_SIZE;
-	tsbsize = num_tsb_entries * sizeof(iopte_t);
 
 	dma_offset = vdma[0];
 
diff --git a/arch/sparc/kernel/pcr.c b/arch/sparc/kernel/pcr.c
index 7c2ced6..8ac23e6 100644
--- a/arch/sparc/kernel/pcr.c
+++ b/arch/sparc/kernel/pcr.c
@@ -81,7 +81,7 @@  static void n2_pcr_write(u64 val)
 	unsigned long ret;
 
 	ret = sun4v_niagara2_setperf(HV_N2_PERF_SPARC_CTL, val);
-	if (val != HV_EOK)
+	if (ret != HV_EOK)
 		write_pcr(val);
 }
 
diff --git a/arch/sparc/kernel/ptrace_64.c b/arch/sparc/kernel/ptrace_64.c
index 9ccc812..96ee50a 100644
--- a/arch/sparc/kernel/ptrace_64.c
+++ b/arch/sparc/kernel/ptrace_64.c
@@ -1086,6 +1086,7 @@  asmlinkage int syscall_trace_enter(struct pt_regs *regs)
 
 asmlinkage void syscall_trace_leave(struct pt_regs *regs)
 {
+#ifdef CONFIG_AUDITSYSCALL
 	if (unlikely(current->audit_context)) {
 		unsigned long tstate = regs->tstate;
 		int result = AUDITSC_SUCCESS;
@@ -1095,7 +1096,7 @@  asmlinkage void syscall_trace_leave(struct pt_regs *regs)
 
 		audit_syscall_exit(result, regs->u_regs[UREG_I0]);
 	}
-
+#endif
 	if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
 		trace_sys_exit(regs, regs->u_regs[UREG_G1]);
 
diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c
index 555a76d..3e94a8c 100644
--- a/arch/sparc/kernel/smp_64.c
+++ b/arch/sparc/kernel/smp_64.c
@@ -189,7 +189,7 @@  static inline long get_delta (long *rt, long *master)
 void smp_synchronize_tick_client(void)
 {
 	long i, delta, adj, adjust_latency = 0, done = 0;
-	unsigned long flags, rt, master_time_stamp, bound;
+	unsigned long flags, rt, master_time_stamp;
 #if DEBUG_TICK_SYNC
 	struct {
 		long rt;	/* roundtrip time */
@@ -208,10 +208,8 @@  void smp_synchronize_tick_client(void)
 	{
 		for (i = 0; i < NUM_ROUNDS; i++) {
 			delta = get_delta(&rt, &master_time_stamp);
-			if (delta == 0) {
+			if (delta == 0)
 				done = 1;	/* let's lock on to this... */
-				bound = rt;
-			}
 
 			if (!done) {
 				if (i > 0) {
@@ -933,13 +931,12 @@  void smp_flush_dcache_page_impl(struct page *page, int cpu)
 void flush_dcache_page_all(struct mm_struct *mm, struct page *page)
 {
 	void *pg_addr;
-	int this_cpu;
 	u64 data0;
 
 	if (tlb_type == hypervisor)
 		return;
 
-	this_cpu = get_cpu();
+	preempt_disable();
 
 #ifdef CONFIG_DEBUG_DCFLUSH
 	atomic_inc(&dcpage_flushes);
@@ -964,7 +961,7 @@  void flush_dcache_page_all(struct mm_struct *mm, struct page *page)
 	}
 	__local_flush_dcache_page(page);
 
-	put_cpu();
+	preempt_enable();
 }
 
 void __irq_entry smp_new_mmu_context_version_client(int irq, struct pt_regs *regs)
diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c
index 42ad2ba..9456472 100644
--- a/arch/sparc/kernel/traps_64.c
+++ b/arch/sparc/kernel/traps_64.c
@@ -2152,7 +2152,7 @@  static void user_instruction_dump(unsigned int __user *pc)
 
 void show_stack(struct task_struct *tsk, unsigned long *_ksp)
 {
-	unsigned long fp, thread_base, ksp;
+	unsigned long fp, ksp;
 	struct thread_info *tp;
 	int count = 0;
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
@@ -2173,7 +2173,6 @@  void show_stack(struct task_struct *tsk, unsigned long *_ksp)
 		flushw_all();
 
 	fp = ksp + STACK_BIAS;
-	thread_base = (unsigned long) tp;
 
 	printk("Call Trace:\n");
 	do {