Re: svn commit: r312792 - in head/sys/arm: arm include
Thank you. On Thu, Jan 26, 2017 at 6:23 AM, Jason A. Harmening wrote: > Author: jah > Date: Thu Jan 26 05:23:33 2017 > New Revision: 312792 > URL: https://svnweb.freebsd.org/changeset/base/312792 > > Log: > Further cleanup of per-CPU armv6 pmap data: > > - Replace pcpu_find(curcpu) with get_pcpu(), which is much > more direct. > > - Remove armv4 pcpu fields which I added in r286296 but never > needed to use. > > - armv6 pc_qmap_addr was leftover from the old armv6 pmap > implementation. Rename it and put it to use in the new one. > > Noted by: skra > Reviewed by: skra > MFC after:1 week > Differential Revision:https://reviews.freebsd.org/D9312 > > Modified: > head/sys/arm/arm/pmap-v6.c > head/sys/arm/include/pcpu.h > [snip] ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312812 - head/sys/arm/mv
Author: wma
Date: Thu Jan 26 11:14:23 2017
New Revision: 312812
URL: https://svnweb.freebsd.org/changeset/base/312812
Log:
Add dummy functions for Marvell SoC's not equipped with AHCI
Commit r312747 ("Setup decoding windows for ARMADA38X") resulted
in build failing for Marvell platforms, which don't have AHCI controller.
This patch provides a fix by adding dummy functions for such cases.
On the occasion rename register dump routine to decode_win_ahci_dump,
in order to avoid confusion.
Submitted by: Marcin Wojtas
Obtained from: Semihalf
Sponsored by: Stormshield
Modified:
head/sys/arm/mv/mv_common.c
Modified: head/sys/arm/mv/mv_common.c
==
--- head/sys/arm/mv/mv_common.c Thu Jan 26 11:07:36 2017(r312811)
+++ head/sys/arm/mv/mv_common.c Thu Jan 26 11:14:23 2017(r312812)
@@ -108,7 +108,7 @@ static void decode_win_usb3_dump(u_long)
static void decode_win_eth_dump(u_long base);
static void decode_win_idma_dump(u_long base);
static void decode_win_xor_dump(u_long base);
-static void decode_win_sata_dump(u_long base);
+static void decode_win_ahci_dump(u_long base);
static int fdt_get_ranges(const char *, void *, int, int *, int *);
#ifdef SOC_MV_ARMADA38X
@@ -141,7 +141,7 @@ static struct soc_node_spec soc_nodes[]
{ "mrvl,ge", &decode_win_eth_setup, &decode_win_eth_dump },
{ "mrvl,usb-ehci", &decode_win_usb_setup, &decode_win_usb_dump },
{ "marvell,armada-380-xhci", &decode_win_usb3_setup,
&decode_win_usb3_dump },
- { "marvell,armada-380-ahci", &decode_win_ahci_setup,
&decode_win_sata_dump },
+ { "marvell,armada-380-ahci", &decode_win_ahci_setup,
&decode_win_ahci_dump },
{ "mrvl,sata", &decode_win_sata_setup, NULL },
{ "mrvl,xor", &decode_win_xor_setup, &decode_win_xor_dump },
{ "mrvl,idma", &decode_win_idma_setup, &decode_win_idma_dump },
@@ -2007,6 +2007,10 @@ decode_win_sata_setup(u_long base)
}
}
+#ifdef SOC_MV_ARMADA38X
+/*
+ * Configure AHCI decoding windows
+ */
static void
decode_win_ahci_setup(u_long base)
{
@@ -2046,7 +2050,7 @@ decode_win_ahci_setup(u_long base)
}
static void
-decode_win_sata_dump(u_long base)
+decode_win_ahci_dump(u_long base)
{
int i;
@@ -2056,6 +2060,22 @@ decode_win_sata_dump(u_long base)
win_sata_sz_read(base,i));
}
+#else
+/*
+ * Provide dummy functions to satisfy the build
+ * for SoC's not equipped with AHCI controller
+ */
+static void
+decode_win_ahci_setup(u_long base)
+{
+}
+
+static void
+decode_win_ahci_dump(u_long base)
+{
+}
+#endif
+
static int
decode_win_sata_valid(void)
{
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r312747 - head/sys/arm/mv
Fixed in https://svnweb.freebsd.org/changeset/base/312812 Regards, Wojtek 2017-01-26 2:56 GMT+01:00 Ed Maste : > On 25 January 2017 at 05:31, Wojciech Macek wrote: > > Author: wma > > Date: Wed Jan 25 10:31:16 2017 > > New Revision: 312747 > > URL: https://svnweb.freebsd.org/changeset/base/312747 > > > > Log: > > Setup decoding windows for ARMADA38X > > > > It is necesarry to open memory windows on internal bus for > > AHCI driver to work correctly. > > Build broken with: > > /scratch/tmp/emaste/freebsd/sys/arm/mv/mv_common.c:2019:3: error: > implicit declaration of function 'win_sata_sz_write' is invalid in C99 > [-Werror,-Wimplicit-function-declaration] > win_sata_sz_write(base, i, 0); > ^ > /scratch/tmp/emaste/freebsd/sys/arm/mv/mv_common.c:2024:25: error: use > of undeclared identifier 'IO_WIN_ATTR_SHIFT' > cr = (ddr_attr(i) << IO_WIN_ATTR_SHIFT) | > ^ > ... > ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312813 - in head/sys: arm/mv dev/fdt
Author: andrew
Date: Thu Jan 26 13:04:14 2017
New Revision: 312813
URL: https://svnweb.freebsd.org/changeset/base/312813
Log:
Make fdt_pm_mask_table internal to the Marvell code, it's unued anywhere
else.
Sponsored by: ABT Systems Ltd
Modified:
head/sys/arm/mv/mv_common.c
head/sys/dev/fdt/fdt_common.h
Modified: head/sys/arm/mv/mv_common.c
==
--- head/sys/arm/mv/mv_common.c Thu Jan 26 11:14:23 2017(r312812)
+++ head/sys/arm/mv/mv_common.c Thu Jan 26 13:04:14 2017(r312813)
@@ -149,7 +149,12 @@ static struct soc_node_spec soc_nodes[]
{ NULL, NULL, NULL },
};
-struct fdt_pm_mask_entry fdt_pm_mask_table[] = {
+struct fdt_pm_mask_entry {
+ char*compat;
+ uint32_tmask;
+};
+
+static struct fdt_pm_mask_entry fdt_pm_mask_table[] = {
{ "mrvl,ge",CPU_PM_CTRL_GE(0) },
{ "mrvl,ge",CPU_PM_CTRL_GE(1) },
{ "mrvl,usb-ehci", CPU_PM_CTRL_USB(0) },
Modified: head/sys/dev/fdt/fdt_common.h
==
--- head/sys/dev/fdt/fdt_common.h Thu Jan 26 11:14:23 2017
(r312812)
+++ head/sys/dev/fdt/fdt_common.h Thu Jan 26 13:04:14 2017
(r312813)
@@ -71,12 +71,6 @@ extern vm_paddr_t fdt_immr_pa;
extern vm_offset_t fdt_immr_va;
extern vm_offset_t fdt_immr_size;
-struct fdt_pm_mask_entry {
- char*compat;
- uint32_tmask;
-};
-extern struct fdt_pm_mask_entry fdt_pm_mask_table[];
-
#if defined(FDT_DTB_STATIC)
extern u_char fdt_static_dtb;
#endif
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312814 - head/sys/sys
Author: sbruno Date: Thu Jan 26 13:46:47 2017 New Revision: 312814 URL: https://svnweb.freebsd.org/changeset/base/312814 Log: Shoot a couple of style bugs down in the macro declarations. Submitted by: bde Modified: head/sys/sys/gtaskqueue.h Modified: head/sys/sys/gtaskqueue.h == --- head/sys/sys/gtaskqueue.h Thu Jan 26 13:04:14 2017(r312813) +++ head/sys/sys/gtaskqueue.h Thu Jan 26 13:46:47 2017(r312814) @@ -80,7 +80,6 @@ int taskqgroup_adjust(struct taskqgroup #define TASKQGROUP_DECLARE(name) \ extern struct taskqgroup *qgroup_##name - #ifdef EARLY_AP_STARTUP #define TASKQGROUP_DEFINE(name, cnt, stride) \ \ @@ -94,8 +93,7 @@ taskqgroup_define_##name(void *arg) } \ \ SYSINIT(taskqgroup_##name, SI_SUB_INIT_IF, SI_ORDER_FIRST, \ - taskqgroup_define_##name, NULL) \ - + taskqgroup_define_##name, NULL) #else /* !EARLY_AP_STARTUP */ #define TASKQGROUP_DEFINE(name, cnt, stride) \ \ @@ -117,8 +115,7 @@ taskqgroup_adjust_##name(void *arg) } \ \ SYSINIT(taskqgroup_adj_##name, SI_SUB_SMP, SI_ORDER_ANY, \ - taskqgroup_adjust_##name, NULL) \ - + taskqgroup_adjust_##name, NULL) #endif /* EARLY_AP_STARTUP */ TASKQGROUP_DECLARE(net); ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312815 - head/sys/kern
Author: sbruno
Date: Thu Jan 26 13:48:45 2017
New Revision: 312815
URL: https://svnweb.freebsd.org/changeset/base/312815
Log:
A few more style bugs lying around in here.
Submitted by: bde
Modified:
head/sys/kern/subr_gtaskqueue.c
Modified: head/sys/kern/subr_gtaskqueue.c
==
--- head/sys/kern/subr_gtaskqueue.c Thu Jan 26 13:46:47 2017
(r312814)
+++ head/sys/kern/subr_gtaskqueue.c Thu Jan 26 13:48:45 2017
(r312815)
@@ -630,6 +630,7 @@ taskqgroup_find(struct taskqgroup *qgrou
return (idx);
}
+
/*
* smp_started is unusable since it is not set for UP kernels or even for
* SMP kernels when there is 1 CPU. This is usually handled by adding a
@@ -643,6 +644,7 @@ taskqgroup_find(struct taskqgroup *qgrou
* SI_ORDER_ANY and unclearly after the CPUs are started. It would be
* simpler for adjustment to pass a flag indicating if it is delayed.
*/
+
static int tqg_smp_started;
static void
@@ -670,7 +672,7 @@ taskqgroup_attach(struct taskqgroup *qgr
qgroup->tqg_queue[qid].tgc_cnt++;
LIST_INSERT_HEAD(&qgroup->tqg_queue[qid].tgc_tasks, gtask, gt_list);
gtask->gt_taskqueue = qgroup->tqg_queue[qid].tgc_taskq;
- if (irq != -1 && tqg_smp_started ) {
+ if (irq != -1 && tqg_smp_started) {
gtask->gt_cpu = qgroup->tqg_queue[qid].tgc_cpu;
CPU_ZERO(&mask);
CPU_SET(qgroup->tqg_queue[qid].tgc_cpu, &mask);
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312816 - head/sys/net
Author: sbruno Date: Thu Jan 26 13:50:09 2017 New Revision: 312816 URL: https://svnweb.freebsd.org/changeset/base/312816 Log: Minor style annoyance. Submitted by: bde Modified: head/sys/net/iflib.c Modified: head/sys/net/iflib.c == --- head/sys/net/iflib.cThu Jan 26 13:48:45 2017(r312815) +++ head/sys/net/iflib.cThu Jan 26 13:50:09 2017(r312816) @@ -3755,6 +3755,7 @@ iflib_device_register(device_t dev, void device_printf(dev, "qset structure setup failed %d\n", err); goto fail_queues; } + /* * Group taskqueues aren't properly set up until SMP is started, * so we disable interrupts until we can handle them post ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312818 - head/share/man/man9
Author: obrien Date: Thu Jan 26 16:36:12 2017 New Revision: 312818 URL: https://svnweb.freebsd.org/changeset/base/312818 Log: Correct grammar. Modified: head/share/man/man9/printf.9 Modified: head/share/man/man9/printf.9 == --- head/share/man/man9/printf.9Thu Jan 26 15:37:53 2017 (r312817) +++ head/share/man/man9/printf.9Thu Jan 26 16:36:12 2017 (r312818) @@ -111,7 +111,7 @@ It requires two arguments: a pointer and a .Vt "char *" string. -The memory pointed to be the pointer is output in hexadecimal one byte at +The memory pointed to by the pointer is output in hexadecimal one byte at a time. The string is used as a delimiter between individual bytes. If present, a width directive will specify the number of bytes to display. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r311993 - head/sys/dev/nand
Should this be merged to 11-stable also? I got a build failure on this with 11/arm. Ronald. On Thu, 12 Jan 2017 19:05:12 +0100, Alexander Kabaev wrote: Author: kan Date: Thu Jan 12 18:05:12 2017 New Revision: 311993 URL: https://svnweb.freebsd.org/changeset/base/311993 Log: Fix typo in r311971. Reported by: ohartmann at walstatt.org Modified: head/sys/dev/nand/nand_geom.c Modified: head/sys/dev/nand/nand_geom.c == --- head/sys/dev/nand/nand_geom.c Thu Jan 12 17:54:55 2017 (r311992) +++ head/sys/dev/nand/nand_geom.c Thu Jan 12 18:05:12 2017 (r311993) @@ -416,7 +416,7 @@ create_geom_disk(struct nand_chip *chip) snprintf(rdisk->d_ident, sizeof(rdisk->d_ident), "nand_raw: Man:0x%02x Dev:0x%02x", chip->id.man_id, chip->id.dev_id); - disk->d_rotation_rate = DISK_RR_NON_ROTATING; + rdisk->d_rotation_rate = DISK_RR_NON_ROTATING; disk_create(rdisk, DISK_VERSION); ___ svn-src-...@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-all To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org" ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312820 - head/sys/modules
Author: emaste
Date: Thu Jan 26 17:59:54 2017
New Revision: 312820
URL: https://svnweb.freebsd.org/changeset/base/312820
Log:
Disconnect netfpga10g module from the build
It only builds with the non-default DEVICE_POLLING option.
Approved by: bz
Modified:
head/sys/modules/Makefile
Modified: head/sys/modules/Makefile
==
--- head/sys/modules/Makefile Thu Jan 26 16:38:53 2017(r312819)
+++ head/sys/modules/Makefile Thu Jan 26 17:59:54 2017(r312820)
@@ -265,7 +265,6 @@ SUBDIR= \
${_nctgpio} \
${_ncv} \
${_ndis} \
- netfpga10g \
${_netgraph} \
${_nfe} \
nfscl \
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312821 - in head/sys/modules: . usb
Author: emaste
Date: Thu Jan 26 18:05:31 2017
New Revision: 312821
URL: https://svnweb.freebsd.org/changeset/base/312821
Log:
mips: exclude modules that fail to build
Modified:
head/sys/modules/Makefile
head/sys/modules/usb/Makefile
Modified: head/sys/modules/Makefile
==
--- head/sys/modules/Makefile Thu Jan 26 17:59:54 2017(r312820)
+++ head/sys/modules/Makefile Thu Jan 26 18:05:31 2017(r312821)
@@ -503,13 +503,16 @@ _bce= bce
_fatm= fatm
_fxp= fxp
_ispfw=ispfw
+_sf= sf
+_ti= ti
+_txp= txp
+
+.if ${MACHINE_CPUARCH} != "mips"
_mwlfw=mwlfw
_otusfw= otusfw
_ralfw=ralfw
_rtwnfw= rtwnfw
-_sf= sf
-_ti= ti
-_txp= txp
+.endif
.endif
.if ${MK_SOURCELESS_UCODE} != "no" && ${MACHINE_CPUARCH} != "arm" && \
Modified: head/sys/modules/usb/Makefile
==
--- head/sys/modules/usb/Makefile Thu Jan 26 17:59:54 2017
(r312820)
+++ head/sys/modules/usb/Makefile Thu Jan 26 18:05:31 2017
(r312821)
@@ -70,9 +70,12 @@ _uath= uath
_zyd= zyd
_kue= kue
_run= run
-_runfw=runfw
_rsu= rsu
+
+.if ${MACHINE_CPUARCH} != "mips"
_rsufw=rsufw
+_runfw=runfw
+.endif
.endif
.if ${MACHINE_CPUARCH} == "amd64"
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312822 - head/sys/mips/conf
Author: emaste Date: Thu Jan 26 18:18:35 2017 New Revision: 312822 URL: https://svnweb.freebsd.org/changeset/base/312822 Log: Enable modules in the MIPS ERL kernel config As reported by cperciva[1], modules are beneficial for typical Edgerouter Lite use cases. [1] http://www.daemonology.net/blog/2016-01-10-FreeBSD-EdgeRouter-Lite.html Modified: head/sys/mips/conf/ERL Modified: head/sys/mips/conf/ERL == --- head/sys/mips/conf/ERL Thu Jan 26 18:05:31 2017(r312821) +++ head/sys/mips/conf/ERL Thu Jan 26 18:18:35 2017(r312822) @@ -24,8 +24,6 @@ ident ERL makeoptionsARCH_FLAGS="-march=octeon -mabi=64" makeoptionsLDSCRIPT_NAME=ldscript.mips.octeon1 -# Don't build any modules yet. -makeoptionsMODULES_OVERRIDE="" makeoptionsKERNLOADADDR=0x8010 # We don't need to build a trampolined version of the kernel. ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312825 - in head: crypto/openssl crypto/openssl/apps crypto/openssl/crypto crypto/openssl/crypto/aes/asm crypto/openssl/crypto/asn1 crypto/openssl/crypto/bn crypto/openssl/crypto/bn/as...
Author: jkim Date: Thu Jan 26 19:10:29 2017 New Revision: 312825 URL: https://svnweb.freebsd.org/changeset/base/312825 Log: Merge OpenSSL 1.0.2k. Modified: head/crypto/openssl/CHANGES head/crypto/openssl/CONTRIBUTING head/crypto/openssl/Configure head/crypto/openssl/INSTALL head/crypto/openssl/Makefile head/crypto/openssl/Makefile.org head/crypto/openssl/NEWS head/crypto/openssl/README head/crypto/openssl/apps/apps.c head/crypto/openssl/apps/apps.h head/crypto/openssl/apps/ca.c head/crypto/openssl/apps/cms.c head/crypto/openssl/apps/dgst.c head/crypto/openssl/apps/dh.c head/crypto/openssl/apps/dhparam.c head/crypto/openssl/apps/dsa.c head/crypto/openssl/apps/dsaparam.c head/crypto/openssl/apps/ec.c head/crypto/openssl/apps/ecparam.c head/crypto/openssl/apps/enc.c head/crypto/openssl/apps/gendh.c head/crypto/openssl/apps/gendsa.c head/crypto/openssl/apps/genpkey.c head/crypto/openssl/apps/genrsa.c head/crypto/openssl/apps/pkcs12.c head/crypto/openssl/apps/pkcs7.c head/crypto/openssl/apps/pkcs8.c head/crypto/openssl/apps/pkey.c head/crypto/openssl/apps/pkeyparam.c head/crypto/openssl/apps/pkeyutl.c head/crypto/openssl/apps/prime.c head/crypto/openssl/apps/rand.c head/crypto/openssl/apps/req.c head/crypto/openssl/apps/rsa.c head/crypto/openssl/apps/rsautl.c head/crypto/openssl/apps/s_cb.c head/crypto/openssl/apps/s_client.c head/crypto/openssl/apps/s_server.c head/crypto/openssl/apps/smime.c head/crypto/openssl/apps/speed.c head/crypto/openssl/apps/spkac.c head/crypto/openssl/apps/srp.c head/crypto/openssl/apps/verify.c head/crypto/openssl/apps/x509.c head/crypto/openssl/crypto/aes/asm/aes-s390x.pl head/crypto/openssl/crypto/asn1/p5_pbev2.c head/crypto/openssl/crypto/asn1/x_crl.c head/crypto/openssl/crypto/bn/asm/x86_64-mont.pl head/crypto/openssl/crypto/bn/asm/x86_64-mont5.pl head/crypto/openssl/crypto/bn/bn_exp.c head/crypto/openssl/crypto/bn/bn_mul.c head/crypto/openssl/crypto/bn/bn_prime.c head/crypto/openssl/crypto/bn/bn_sqr.c head/crypto/openssl/crypto/cms/cms_kari.c head/crypto/openssl/crypto/dh/dh_key.c head/crypto/openssl/crypto/dsa/dsa_pmeth.c head/crypto/openssl/crypto/ec/ec2_mult.c head/crypto/openssl/crypto/ecdh/ech_ossl.c head/crypto/openssl/crypto/err/err.c head/crypto/openssl/crypto/evp/e_aes.c head/crypto/openssl/crypto/evp/e_rc4_hmac_md5.c head/crypto/openssl/crypto/evp/evp.h head/crypto/openssl/crypto/evp/evp_err.c head/crypto/openssl/crypto/evp/pmeth_fn.c head/crypto/openssl/crypto/evp/pmeth_lib.c head/crypto/openssl/crypto/modes/ctr128.c head/crypto/openssl/crypto/opensslv.h head/crypto/openssl/crypto/perlasm/x86_64-xlate.pl head/crypto/openssl/crypto/rsa/rsa_gen.c head/crypto/openssl/crypto/rsa/rsa_oaep.c head/crypto/openssl/crypto/rsa/rsa_pmeth.c head/crypto/openssl/crypto/s390xcap.c head/crypto/openssl/crypto/ui/ui_lib.c head/crypto/openssl/crypto/ui/ui_openssl.c head/crypto/openssl/doc/apps/ocsp.pod head/crypto/openssl/doc/crypto/EVP_DigestSignInit.pod head/crypto/openssl/doc/crypto/EVP_DigestVerifyInit.pod head/crypto/openssl/doc/crypto/RSA_generate_key.pod head/crypto/openssl/doc/crypto/X509_NAME_get_index_by_NID.pod head/crypto/openssl/doc/crypto/X509_NAME_print_ex.pod head/crypto/openssl/doc/ssl/SSL_CTX_set_session_cache_mode.pod head/crypto/openssl/doc/ssl/SSL_get_error.pod head/crypto/openssl/doc/ssl/SSL_read.pod head/crypto/openssl/doc/ssl/SSL_write.pod head/crypto/openssl/engines/ccgost/Makefile head/crypto/openssl/ssl/bad_dtls_test.c head/crypto/openssl/ssl/s23_pkt.c head/crypto/openssl/ssl/s2_lib.c head/crypto/openssl/ssl/s2_pkt.c head/crypto/openssl/ssl/s3_clnt.c head/crypto/openssl/ssl/s3_pkt.c head/crypto/openssl/ssl/s3_srvr.c head/crypto/openssl/ssl/ssl_cert.c head/crypto/openssl/ssl/ssl_err.c head/crypto/openssl/ssl/ssl_lib.c head/crypto/openssl/ssl/ssl_locl.h head/crypto/openssl/ssl/ssl_sess.c head/crypto/openssl/ssl/t1_lib.c head/crypto/openssl/util/domd head/crypto/openssl/util/mklink.pl head/secure/lib/libcrypto/Makefile.inc head/secure/lib/libcrypto/amd64/x86_64-mont.S head/secure/lib/libcrypto/amd64/x86_64-mont5.S head/secure/lib/libcrypto/man/ASN1_OBJECT_new.3 head/secure/lib/libcrypto/man/ASN1_STRING_length.3 head/secure/lib/libcrypto/man/ASN1_STRING_new.3 head/secure/lib/libcrypto/man/ASN1_STRING_print_ex.3 head/secure/lib/libcrypto/man/ASN1_TIME_set.3 head/secure/lib/libcrypto/man/ASN1_generate_nconf.3 head/secure/lib/libcrypto/man/BIO_ctrl.3 head/secure/lib/libcrypto/man/BIO_f_base64.3 head/secure/lib/libcrypto/man/BIO_f_buffer.3 head/secure/lib/libcrypto/man/BIO_f_cipher.3 head/secure/lib/libcrypto/man/BIO_f_md.3 head/secure/lib/libcrypto/man/BIO_f_null.3 head/secure/lib/libcrypto/man/BIO_f_ssl.3 head/secure/lib/libcrypto/man/BIO_find_type.3 head/secure/lib/libcrypto/man/BIO_new.3 head/secure/lib/libcrypto/man/BIO_new_CM
svn commit: r312827 - head/sys/cam
Author: scottl
Date: Thu Jan 26 20:08:58 2017
New Revision: 312827
URL: https://svnweb.freebsd.org/changeset/base/312827
Log:
Refactor xpt_print_path, xpt_print, and xpt_path_string. Implement
all of them in terms of an sbuf-based back-end, xpt_path_sbuf. This
unifies the implementation, but more importantly it stops the output
fropm being split between 4 or more invocations of printf. The
multiple invocations cause interleaving of the messages on the
console during boot, making the output of disk discovery often
unintelligible. This change helps a lot, but more work is needed.
Reviewed by: ken, mav
Sponsored by: Netflix
Modified:
head/sys/cam/cam_xpt.c
head/sys/cam/cam_xpt.h
Modified: head/sys/cam/cam_xpt.c
==
--- head/sys/cam/cam_xpt.c Thu Jan 26 19:14:14 2017(r312826)
+++ head/sys/cam/cam_xpt.c Thu Jan 26 20:08:58 2017(r312827)
@@ -3697,33 +3697,14 @@ xpt_path_comp_dev(struct cam_path *path,
void
xpt_print_path(struct cam_path *path)
{
+ struct sbuf sb;
+ char buffer[XPT_PRINT_LEN];
- if (path == NULL)
- printf("(nopath): ");
- else {
- if (path->periph != NULL)
- printf("(%s%d:", path->periph->periph_name,
- path->periph->unit_number);
- else
- printf("(noperiph:");
-
- if (path->bus != NULL)
- printf("%s%d:%d:", path->bus->sim->sim_name,
- path->bus->sim->unit_number,
- path->bus->sim->bus_id);
- else
- printf("nobus:");
-
- if (path->target != NULL)
- printf("%d:", path->target->target_id);
- else
- printf("X:");
-
- if (path->device != NULL)
- printf("%jx): ", (uintmax_t)path->device->lun_id);
- else
- printf("X): ");
- }
+ sbuf_new(&sb, buffer, XPT_PRINT_LEN, SBUF_FIXEDLEN);
+ xpt_path_sbuf(path, &sb);
+ sbuf_finish(&sb);
+ printf("%s", sbuf_data(&sb));
+ sbuf_delete(&sb);
}
void
@@ -3745,49 +3726,66 @@ void
xpt_print(struct cam_path *path, const char *fmt, ...)
{
va_list ap;
- xpt_print_path(path);
+ struct sbuf sb;
+ char buffer[XPT_PRINT_MAXLEN];
+
+ sbuf_new(&sb, buffer, XPT_PRINT_MAXLEN, SBUF_FIXEDLEN);
+
+ xpt_path_sbuf(path, &sb);
va_start(ap, fmt);
- vprintf(fmt, ap);
+ sbuf_vprintf(&sb, fmt, ap);
va_end(ap);
+
+ sbuf_finish(&sb);
+ printf("%s", sbuf_data(&sb));
+ sbuf_delete(&sb);
}
int
xpt_path_string(struct cam_path *path, char *str, size_t str_len)
{
struct sbuf sb;
+ int len;
sbuf_new(&sb, str, str_len, 0);
+ len = xpt_path_sbuf(path, &sb);
+ sbuf_finish(&sb);
+ return (len);
+}
+
+int
+xpt_path_sbuf(struct cam_path *path, struct sbuf *sb)
+{
if (path == NULL)
- sbuf_printf(&sb, "(nopath): ");
+ sbuf_printf(sb, "(nopath): ");
else {
if (path->periph != NULL)
- sbuf_printf(&sb, "(%s%d:", path->periph->periph_name,
+ sbuf_printf(sb, "(%s%d:", path->periph->periph_name,
path->periph->unit_number);
else
- sbuf_printf(&sb, "(noperiph:");
+ sbuf_printf(sb, "(noperiph:");
if (path->bus != NULL)
- sbuf_printf(&sb, "%s%d:%d:", path->bus->sim->sim_name,
+ sbuf_printf(sb, "%s%d:%d:", path->bus->sim->sim_name,
path->bus->sim->unit_number,
path->bus->sim->bus_id);
else
- sbuf_printf(&sb, "nobus:");
+ sbuf_printf(sb, "nobus:");
if (path->target != NULL)
- sbuf_printf(&sb, "%d:", path->target->target_id);
+ sbuf_printf(sb, "%d:", path->target->target_id);
else
- sbuf_printf(&sb, "X:");
+ sbuf_printf(sb, "X:");
if (path->device != NULL)
- sbuf_printf(&sb, "%jx): ",
+ sbuf_printf(sb, "%jx): ",
(uintmax_t)path->device->lun_id);
else
- sbuf_printf(&sb, "X): ");
+ sbuf_printf(sb, "X): ");
}
- sbuf_finish(&sb);
- return(sbuf_len(&sb));
+ return(sbuf_len(sb));
}
path_id_t
Modified: head/sys/cam/cam_xpt.h
==
--- head/sys/cam/cam_xpt.
svn commit: r312830 - head/sys/cam
Author: scottl
Date: Thu Jan 26 20:18:03 2017
New Revision: 312830
URL: https://svnweb.freebsd.org/changeset/base/312830
Log:
Fix a development mis-merge from r312827
Sponsored by: Netflix
Modified:
head/sys/cam/cam_xpt.c
Modified: head/sys/cam/cam_xpt.c
==
--- head/sys/cam/cam_xpt.c Thu Jan 26 20:15:14 2017(r312829)
+++ head/sys/cam/cam_xpt.c Thu Jan 26 20:18:03 2017(r312830)
@@ -3727,9 +3727,9 @@ xpt_print(struct cam_path *path, const c
{
va_list ap;
struct sbuf sb;
- char buffer[XPT_PRINT_MAXLEN];
+ char buffer[XPT_PRINT_LEN];
- sbuf_new(&sb, buffer, XPT_PRINT_MAXLEN, SBUF_FIXEDLEN);
+ sbuf_new(&sb, buffer, XPT_PRINT_LEN, SBUF_FIXEDLEN);
xpt_path_sbuf(path, &sb);
va_start(ap, fmt);
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312831 - in head/contrib/llvm: include/llvm/Analysis lib/Analysis
Author: dim
Date: Thu Jan 26 20:18:28 2017
New Revision: 312831
URL: https://svnweb.freebsd.org/changeset/base/312831
Log:
Revert r312765 for now, since it causes assertions when building
lang/spidermonkey24.
Reported by: antoine
PR: 215649
Modified:
head/contrib/llvm/include/llvm/Analysis/ScalarEvolution.h
head/contrib/llvm/include/llvm/Analysis/ScalarEvolutionExpander.h
head/contrib/llvm/lib/Analysis/ScalarEvolution.cpp
head/contrib/llvm/lib/Analysis/ScalarEvolutionExpander.cpp
Modified: head/contrib/llvm/include/llvm/Analysis/ScalarEvolution.h
==
--- head/contrib/llvm/include/llvm/Analysis/ScalarEvolution.h Thu Jan 26
20:18:03 2017(r312830)
+++ head/contrib/llvm/include/llvm/Analysis/ScalarEvolution.h Thu Jan 26
20:18:28 2017(r312831)
@@ -495,29 +495,10 @@ namespace llvm {
/// The typedef for ExprValueMap.
///
-typedef std::pair ValueOffsetPair;
-typedef DenseMap>
ExprValueMapType;
+typedef DenseMap> ExprValueMapType;
/// ExprValueMap -- This map records the original values from which
/// the SCEV expr is generated from.
-///
-/// We want to represent the mapping as SCEV -> ValueOffsetPair instead
-/// of SCEV -> Value:
-/// Suppose we know S1 expands to V1, and
-/// S1 = S2 + C_a
-/// S3 = S2 + C_b
-/// where C_a and C_b are different SCEVConstants. Then we'd like to
-/// expand S3 as V1 - C_a + C_b instead of expanding S2 literally.
-/// It is helpful when S2 is a complex SCEV expr.
-///
-/// In order to do that, we represent ExprValueMap as a mapping from
-/// SCEV to ValueOffsetPair. We will save both S1->{V1, 0} and
-/// S2->{V1, C_a} into the map when we create SCEV for V1. When S3
-/// is expanded, it will first expand S2 to V1 - C_a because of
-/// S2->{V1, C_a} in the map, then expand S3 to V1 - C_a + C_b.
-///
-/// Note: S->{V, Offset} in the ExprValueMap means S can be expanded
-/// to V - Offset.
ExprValueMapType ExprValueMap;
/// The typedef for ValueExprMap.
@@ -1200,7 +1181,7 @@ namespace llvm {
bool containsAddRecurrence(const SCEV *S);
/// Return the Value set from which the SCEV expr is generated.
-SetVector *getSCEVValues(const SCEV *S);
+SetVector *getSCEVValues(const SCEV *S);
/// Erase Value from ValueExprMap and ExprValueMap.
void eraseValueFromMap(Value *V);
Modified: head/contrib/llvm/include/llvm/Analysis/ScalarEvolutionExpander.h
==
--- head/contrib/llvm/include/llvm/Analysis/ScalarEvolutionExpander.h Thu Jan
26 20:18:03 2017(r312830)
+++ head/contrib/llvm/include/llvm/Analysis/ScalarEvolutionExpander.h Thu Jan
26 20:18:28 2017(r312831)
@@ -325,8 +325,7 @@ namespace llvm {
PointerType *PTy, Type *Ty, Value *V);
/// \brief Find a previous Value in ExprValueMap for expand.
-ScalarEvolution::ValueOffsetPair
-FindValueInExprValueMap(const SCEV *S, const Instruction *InsertPt);
+Value *FindValueInExprValueMap(const SCEV *S, const Instruction *InsertPt);
Value *expand(const SCEV *S);
Modified: head/contrib/llvm/lib/Analysis/ScalarEvolution.cpp
==
--- head/contrib/llvm/lib/Analysis/ScalarEvolution.cpp Thu Jan 26 20:18:03
2017(r312830)
+++ head/contrib/llvm/lib/Analysis/ScalarEvolution.cpp Thu Jan 26 20:18:28
2017(r312831)
@@ -3378,28 +3378,8 @@ bool ScalarEvolution::containsAddRecurre
return F.FoundOne;
}
-/// Try to split a SCEVAddExpr into a pair of {SCEV, ConstantInt}.
-/// If \p S is a SCEVAddExpr and is composed of a sub SCEV S' and an
-/// offset I, then return {S', I}, else return {\p S, nullptr}.
-static std::pair splitAddExpr(const SCEV *S) {
- const auto *Add = dyn_cast(S);
- if (!Add)
-return {S, nullptr};
-
- if (Add->getNumOperands() != 2)
-return {S, nullptr};
-
- auto *ConstOp = dyn_cast(Add->getOperand(0));
- if (!ConstOp)
-return {S, nullptr};
-
- return {Add->getOperand(1), ConstOp->getValue()};
-}
-
-/// Return the ValueOffsetPair set for \p S. \p S can be represented
-/// by the value and offset from any ValueOffsetPair in the set.
-SetVector *
-ScalarEvolution::getSCEVValues(const SCEV *S) {
+/// Return the Value set from S.
+SetVector *ScalarEvolution::getSCEVValues(const SCEV *S) {
ExprValueMapType::iterator SI = ExprValueMap.find_as(S);
if (SI == ExprValueMap.end())
return nullptr;
@@ -3407,31 +3387,24 @@ ScalarEvolution::getSCEVValues(const SCE
if (VerifySCEVMap) {
// Check there is no dangling Value in the set returned.
for (const auto &VE : SI->second)
- assert(ValueExprMap.count(VE.first));
+ assert(ValueExprMap.count(VE));
}
#endif
return &SI->second;
}
-/// Era
svn commit: r312832 - in head/contrib/llvm: include/llvm/Analysis lib/Analysis lib/Transforms/Scalar
Author: dim
Date: Thu Jan 26 20:39:43 2017
New Revision: 312832
URL: https://svnweb.freebsd.org/changeset/base/312832
Log:
Pull in r278160 from upstream llvm trunk (by Wei Mi):
Recommit "Use ValueOffsetPair to enhance value reuse during SCEV
expansion".
The fix for PR28705 will be committed consecutively.
In D12090, the ExprValueMap was added to reuse existing value during
SCEV expansion. However, const folding and sext/zext distribution can
make the reuse still difficult.
A simplified case is: suppose we know S1 expands to V1 in
ExprValueMap, and
S1 = S2 + C_a
S3 = S2 + C_b
where C_a and C_b are different SCEVConstants. Then we'd like to
expand S3 as V1 - C_a + C_b instead of expanding S2 literally. It is
helpful when S2 is a complex SCEV expr and S2 has no entry in
ExprValueMap, which is usually caused by the fact that S3 is
generated from S1 after const folding.
In order to do that, we represent ExprValueMap as a mapping from SCEV
to ValueOffsetPair. We will save both S1->{V1, 0} and S2->{V1, C_a}
into the ExprValueMap when we create SCEV for V1. When S3 is
expanded, it will first expand S2 to V1 - C_a because of S2->{V1,
C_a} in the map, then expand S3 to V1 - C_a + C_b.
Differential Revision: https://reviews.llvm.org/D21313
Pull in r278161 from upstream llvm trunk (by Wei Mi):
Fix the runtime error caused by "Use ValueOffsetPair to enhance value
reuse during SCEV expansion".
The patch is to fix the bug in PR28705. It was caused by setting
wrong return value for SCEVExpander::findExistingExpansion. The
return values of findExistingExpansion have different meanings when
the function is used in different ways so it is easy to make mistake.
The fix creates two new interfaces to replace
SCEVExpander::findExistingExpansion, and specifies where each
interface is expected to be used.
Differential Revision: https://reviews.llvm.org/D22942
Pull in r281439 from upstream llvm trunk (by Wei Mi):
Create a getelementptr instead of sub expr for ValueOffsetPair if the
value is a pointer.
This patch is to fix PR30213. When expanding an expr based on
ValueOffsetPair, if the value is of pointer type, we can only create
a getelementptr instead of sub expr.
Differential Revision: https://reviews.llvm.org/D24088
This should fix assertion failures when building OpenCV >= 3.1, and also
allow building lang/spidermonkey24 without any further assertions.
PR: 215649
MFC after:1 week
Modified:
head/contrib/llvm/include/llvm/Analysis/ScalarEvolution.h
head/contrib/llvm/include/llvm/Analysis/ScalarEvolutionExpander.h
head/contrib/llvm/lib/Analysis/ScalarEvolution.cpp
head/contrib/llvm/lib/Analysis/ScalarEvolutionExpander.cpp
head/contrib/llvm/lib/Transforms/Scalar/IndVarSimplify.cpp
Modified: head/contrib/llvm/include/llvm/Analysis/ScalarEvolution.h
==
--- head/contrib/llvm/include/llvm/Analysis/ScalarEvolution.h Thu Jan 26
20:18:28 2017(r312831)
+++ head/contrib/llvm/include/llvm/Analysis/ScalarEvolution.h Thu Jan 26
20:39:43 2017(r312832)
@@ -495,10 +495,29 @@ namespace llvm {
/// The typedef for ExprValueMap.
///
-typedef DenseMap> ExprValueMapType;
+typedef std::pair ValueOffsetPair;
+typedef DenseMap>
ExprValueMapType;
/// ExprValueMap -- This map records the original values from which
/// the SCEV expr is generated from.
+///
+/// We want to represent the mapping as SCEV -> ValueOffsetPair instead
+/// of SCEV -> Value:
+/// Suppose we know S1 expands to V1, and
+/// S1 = S2 + C_a
+/// S3 = S2 + C_b
+/// where C_a and C_b are different SCEVConstants. Then we'd like to
+/// expand S3 as V1 - C_a + C_b instead of expanding S2 literally.
+/// It is helpful when S2 is a complex SCEV expr.
+///
+/// In order to do that, we represent ExprValueMap as a mapping from
+/// SCEV to ValueOffsetPair. We will save both S1->{V1, 0} and
+/// S2->{V1, C_a} into the map when we create SCEV for V1. When S3
+/// is expanded, it will first expand S2 to V1 - C_a because of
+/// S2->{V1, C_a} in the map, then expand S3 to V1 - C_a + C_b.
+///
+/// Note: S->{V, Offset} in the ExprValueMap means S can be expanded
+/// to V - Offset.
ExprValueMapType ExprValueMap;
/// The typedef for ValueExprMap.
@@ -1181,7 +1200,7 @@ namespace llvm {
bool containsAddRecurrence(const SCEV *S);
/// Return the Value set from which the SCEV expr is generated.
-SetVector *getSCEVValues(const SCEV *S);
+SetVector *getSCEVValues(const SCEV *S);
/// Erase Value from ValueExprMap and ExprValueMap.
void eraseValueFromMap(Value *V);
Modified: head/contrib/llvm/include/llvm/Analysis/ScalarEvol
svn commit: r312833 - head/sys/mips/conf
Author: lidl Date: Thu Jan 26 20:45:04 2017 New Revision: 312833 URL: https://svnweb.freebsd.org/changeset/base/312833 Log: Remove 'options NO_SWAPPING' from ERL configuration file Modified: head/sys/mips/conf/ERL Modified: head/sys/mips/conf/ERL == --- head/sys/mips/conf/ERL Thu Jan 26 20:39:43 2017(r312832) +++ head/sys/mips/conf/ERL Thu Jan 26 20:45:04 2017(r312833) @@ -88,7 +88,6 @@ options MAC # TrustedBSD MAC Framewor #options KDTRACE_FRAME # Ensure frames are compiled in #options KDTRACE_HOOKS # Kernel DTrace hooks optionsINCLUDE_CONFIG_FILE # Include this file in kernel -optionsNO_SWAPPING # Disable support for paging optionsTMPFS # Temporary file system # Debugging for use in -current ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r312770 - in head/sys: net netinet netinet6
On Thu, Jan 26, 2017 at 02:03:05PM +1100, Bruce Evans wrote:
B> On Thu, 26 Jan 2017, Konstantin Belousov wrote:
B>
B> > On Wed, Jan 25, 2017 at 02:20:06PM -0800, Gleb Smirnoff wrote:
B> >> Thanks, Luiz!
B> >>
B> >> One stylistic nit that I missed in review:
B> >>
B> >> L> static int
B> >> L> -in_difaddr_ioctl(caddr_t data, struct ifnet *ifp, struct thread *td)
B> >> L> +in_difaddr_ioctl(u_long cmd, caddr_t data, struct ifnet *ifp, struct
thread *td)
B> >> L> {
B> >> L>const struct ifreq *ifr = (struct ifreq *)data;
B> >> L>const struct sockaddr_in *addr = (const struct sockaddr_in *)
B> >> L> @@ -618,7 +618,8 @@ in_difaddr_ioctl(caddr_t data, struct if
B> >> L>in_ifadown(&ia->ia_ifa, 1);
B> >> L>
B> >> L>if (ia->ia_ifa.ifa_carp)
B> >> L> - (*carp_detach_p)(&ia->ia_ifa);
B> >> L> + (*carp_detach_p)(&ia->ia_ifa,
B> >> L> + (cmd == SIOCDIFADDR) ? false : true);
B> >>
B> >> Can we change the very last line to:
B> >>
B> >> (cmd == SIOCAIFADDR) ? true : false);
B>
B> That is not stylistic, but invert the result. Perhaps you meant to
B> reverse the test to avoid negative logic for the result.
It uses different ioctl value, so it doesn't invert result. Instead
of !SIOCDIFADDR I want more explicit SIOCAIFADDR.
--
Totus tuus, Glebius.
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312852 - head/sys/cam
Author: jkim Date: Fri Jan 27 00:17:07 2017 New Revision: 312852 URL: https://svnweb.freebsd.org/changeset/base/312852 Log: Fix libcam build. It was broken with r312827. Modified: head/sys/cam/cam_xpt.h Modified: head/sys/cam/cam_xpt.h == --- head/sys/cam/cam_xpt.h Thu Jan 26 23:29:30 2017(r312851) +++ head/sys/cam/cam_xpt.h Fri Jan 27 00:17:07 2017(r312852) @@ -32,8 +32,10 @@ #ifndef _CAM_CAM_XPT_H #define _CAM_CAM_XPT_H 1 +#ifdef _KERNEL #include #include "opt_printf.h" +#endif /* Forward Declarations */ union ccb; ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312853 - head/sys/dev/ath
Author: adrian
Date: Fri Jan 27 01:17:00 2017
New Revision: 312853
URL: https://svnweb.freebsd.org/changeset/base/312853
Log:
[ath] fix "doing stuff before wakeup" warning; add comments for ACK/CTS
handling during DFS/PASSIVE channels
* Although the hardware is awake, the power state handling doesn't think so.
So just explicitly wake it up early in setup so ath_hal calls don't
complain.
* We shouldn't be transmitting or ACKing frames during DFS CAC or on passive
channels before we hear a beacon. So, start laying down comments in the
places where this work has to be done.
Note:
* The main bit missing from finishing this particular bit of work is a state
call to transition a VAP from passive to non-passive when a beacon is heard.
CAC is easy, it's an interface state. So, I'll go and add a method to
control
that soon.
Modified:
head/sys/dev/ath/if_ath.c
Modified: head/sys/dev/ath/if_ath.c
==
--- head/sys/dev/ath/if_ath.c Fri Jan 27 00:17:07 2017(r312852)
+++ head/sys/dev/ath/if_ath.c Fri Jan 27 01:17:00 2017(r312853)
@@ -634,6 +634,17 @@ ath_attach(u_int16_t devid, struct ath_s
#endif
/*
+* Force the chip awake during setup, just to keep
+* the HAL/driver power tracking happy.
+*
+* There are some methods (eg ath_hal_setmac())
+* that poke the hardware.
+*/
+ ATH_LOCK(sc);
+ ath_power_setpower(sc, HAL_PM_AWAKE, 1);
+ ATH_UNLOCK(sc);
+
+ /*
* Setup the DMA/EDMA functions based on the current
* hardware support.
*
@@ -1010,6 +1021,28 @@ ath_attach(u_int16_t devid, struct ath_s
sc->sc_rx_lnamixer = ath_hal_hasrxlnamixer(ah);
sc->sc_hasdivcomb = ath_hal_hasdivantcomb(ah);
+ /*
+* Some WB335 cards do not support antenna diversity. Since
+* we use a hardcoded value for AR9565 instead of using the
+* EEPROM/OTP data, remove the combining feature from
+* the HW capabilities bitmap.
+*/
+ /*
+* XXX TODO: check reference driver and ath9k for what to do
+* here for WB335. I think we have to actually disable the
+* LNA div processing in the HAL and instead use the hard
+* coded values; and then use BT diversity.
+*
+* .. but also need to setup MCI too for WB335..
+*/
+#if 0
+ if (sc->sc_pci_devinfo & (ATH9K_PCI_AR9565_1ANT |
ATH9K_PCI_AR9565_2ANT)) {
+ device_printf(sc->sc_dev, "%s: WB335: disabling LNA mixer
diversity\n",
+ __func__);
+ sc->sc_dolnadiv = 0;
+ }
+#endif
+
if (ath_hal_hasfastframes(ah))
ic->ic_caps |= IEEE80211_C_FF;
wmodes = ath_hal_getwirelessmodes(ah);
@@ -1351,6 +1384,7 @@ bad2:
ath_desc_free(sc);
ath_txdma_teardown(sc);
ath_rxdma_teardown(sc);
+
bad:
if (ah)
ath_hal_detach(ah);
@@ -3611,6 +3645,8 @@ ath_mode_init(struct ath_softc *sc)
struct ath_hal *ah = sc->sc_ah;
u_int32_t rfilt;
+ /* XXX power state? */
+
/* configure rx filter */
rfilt = ath_calcrxfilter(sc);
ath_hal_setrxfilter(ah, rfilt);
@@ -5654,6 +5690,56 @@ ath_newstate(struct ieee80211vap *vap, e
*/
IEEE80211_LOCK_ASSERT(ic);
+ /*
+* XXX TODO: if nstate is _S_CAC, then we should disable
+* ACK processing until CAC is completed.
+*/
+
+ /*
+* XXX TODO: if we're on a passive channel, then we should
+* not allow any ACKs or self-generated frames until we hear
+* a beacon. Unfortunately there isn't a notification from
+* net80211 so perhaps we could slot that particular check
+* into the mgmt receive path and just ensure that we clear
+* it on RX of beacons in passive mode (and only clear it
+* once, obviously.)
+*/
+
+ /*
+* XXX TODO: net80211 should be tracking whether channels
+* have heard beacons and are thus considered "OK" for
+* transmitting - and then inform the driver about this
+* state change. That way if we hear an AP go quiet
+* (and nothing else is beaconing on a channel) the
+* channel can go back to being passive until another
+* beacon is heard.
+*/
+
+ /*
+* XXX TODO: if nstate is _S_CAC, then we should disable
+* ACK processing until CAC is completed.
+*/
+
+ /*
+* XXX TODO: if we're on a passive channel, then we should
+* not allow any ACKs or self-generated frames until we hear
+* a beacon. Unfortunately there isn't a notification from
+* net80211 so perhaps we could slot that particular check
+* into the mgmt receive path and just ensure that we clear
+* it on RX of beacons in passi
svn commit: r312854 - head/sys/net80211
Author: adrian Date: Fri Jan 27 01:24:24 2017 New Revision: 312854 URL: https://svnweb.freebsd.org/changeset/base/312854 Log: [net80211] prepare configuration checks for VHT, fragment-offload and seqno-offload. * allocate an ext bit for fragment offload. Some NICs (like the ath10k hardware in native wifi or 802.3 mode) support doing packet fragmentation in firmware/hardware, so we don't have to do it here. * allocate an ext bit for VHT and start using it. Modified: head/sys/net80211/ieee80211_var.h Modified: head/sys/net80211/ieee80211_var.h == --- head/sys/net80211/ieee80211_var.h Fri Jan 27 01:17:00 2017 (r312853) +++ head/sys/net80211/ieee80211_var.h Fri Jan 27 01:24:24 2017 (r312854) @@ -93,10 +93,13 @@ * says that VHT is supported - and then this macro can be * changed. */ -#defineIEEE80211_CONF_VHT(ic) ((ic)->ic_vhtcaps != 0) +#defineIEEE80211_CONF_VHT(ic) \ + ((ic)->ic_flags_ext & IEEE80211_FEXT_VHT) #defineIEEE80211_CONF_SEQNO_OFFLOAD(ic)\ ((ic)->ic_flags_ext & IEEE80211_FEXT_SEQNO_OFFLOAD) +#defineIEEE80211_CONF_FRAG_OFFLOAD(ic) \ + ((ic)->ic_flags_ext & IEEE80211_FEXT_FRAG_OFFLOAD) /* * 802.11 control state is split into a common portion that maps @@ -633,11 +636,14 @@ MALLOC_DECLARE(M_80211_VAP); #defineIEEE80211_FEXT_UNIQMAC 0x0004 /* CONF: user or computed mac */ #defineIEEE80211_FEXT_SCAN_OFFLOAD 0x0008 /* CONF: scan is fully offloaded */ #defineIEEE80211_FEXT_SEQNO_OFFLOAD0x0010 /* CONF: driver does seqno insertion/allocation */ +#defineIEEE80211_FEXT_FRAG_OFFLOAD 0x0020 /* CONF: hardware does 802.11 fragmentation + assignment */ +#defineIEEE80211_FEXT_VHT 0x0040 /* CONF: VHT support */ #defineIEEE80211_FEXT_BITS \ "\20\2INACT\3SCANWAIT\4BGSCAN\5WPS\6TSN\7SCANREQ\10RESUME" \ "\0114ADDR\12NONEPR_PR\13SWBMISS\14DFS\15DOTD\16STATEWAIT\17REINIT" \ - "\20BPF\21WDSLEGACY\22PROBECHAN\23UNIQMAC\24SCAN_OFFLOAD\25SEQNO_OFFLOAD" + "\20BPF\21WDSLEGACY\22PROBECHAN\23UNIQMAC\24SCAN_OFFLOAD\25SEQNO_OFFLOAD" \ + "\26VHT" /* ic_flags_ht/iv_flags_ht */ #defineIEEE80211_FHT_NONHT_PR 0x0001 /* STATUS: non-HT sta present */ ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312855 - in head: . gnu/usr.bin/binutils/ld share/mk tools/build/options usr.bin/clang/lld
Author: emaste
Date: Fri Jan 27 01:59:12 2017
New Revision: 312855
URL: https://svnweb.freebsd.org/changeset/base/312855
Log:
Rename LLD_AS_LD to LLD_IS_LD, for consistency with CLANG_IS_CC
Reported by: Dan McGregor
Added:
head/tools/build/options/WITHOUT_LLD_IS_LD
- copied unchanged from r312854, head/tools/build/options/WITHOUT_LLD_AS_LD
head/tools/build/options/WITH_LLD_IS_LD
- copied unchanged from r312854, head/tools/build/options/WITH_LLD_AS_LD
Deleted:
head/tools/build/options/WITHOUT_LLD_AS_LD
head/tools/build/options/WITH_LLD_AS_LD
Modified:
head/Makefile.inc1
head/UPDATING
head/gnu/usr.bin/binutils/ld/Makefile
head/share/mk/src.opts.mk
head/usr.bin/clang/lld/Makefile
Modified: head/Makefile.inc1
==
--- head/Makefile.inc1 Fri Jan 27 01:24:24 2017(r312854)
+++ head/Makefile.inc1 Fri Jan 27 01:59:12 2017(r312855)
@@ -516,7 +516,7 @@ TMAKE= MAKEOBJDIRPREFIX=${OBJTREE} \
# cross-tools stage
XMAKE= TOOLS_PREFIX=${WORLDTMP} ${BMAKE} \
TARGET=${TARGET} TARGET_ARCH=${TARGET_ARCH} \
- MK_GDB=no MK_TESTS=no MK_LLD_AS_LD=no
+ MK_GDB=no MK_TESTS=no MK_LLD_IS_LD=no
# kernel-tools stage
KTMAKEENV= INSTALL="sh ${.CURDIR}/tools/install.sh" \
Modified: head/UPDATING
==
--- head/UPDATING Fri Jan 27 01:24:24 2017(r312854)
+++ head/UPDATING Fri Jan 27 01:59:12 2017(r312855)
@@ -51,6 +51,10 @@ NOTE TO PEOPLE WHO THINK THAT FreeBSD 12
** SPECIAL WARNING: **
+20170127:
+ The WITH_LLD_AS_LD / WITHOUT_LLD_AS_LD build knobs have been renamed
+ WITH_LLD_IS_LD / WITHOUT_LLD_IS_LD, for consistency with CLANG_IS_CC.
+
20170112:
The EM_MULTIQUEUE kernel configuration option is deprecated now that
the em(4) driver conforms to iflib specifications.
Modified: head/gnu/usr.bin/binutils/ld/Makefile
==
--- head/gnu/usr.bin/binutils/ld/Makefile Fri Jan 27 01:24:24 2017
(r312854)
+++ head/gnu/usr.bin/binutils/ld/Makefile Fri Jan 27 01:59:12 2017
(r312855)
@@ -49,7 +49,7 @@ CLEANFILES+= ldemul-list.h stringify.sed
FILES= ${LDSCRIPTS:S|^|ldscripts/|}
FILESDIR= ${SCRIPTDIR}
-.if ${MK_LLD_AS_LD} == "no"
+.if ${MK_LLD_IS_LD} == "no"
LINKS= ${BINDIR}/ld.bfd ${BINDIR}/ld
.endif
Modified: head/share/mk/src.opts.mk
==
--- head/share/mk/src.opts.mk Fri Jan 27 01:24:24 2017(r312854)
+++ head/share/mk/src.opts.mk Fri Jan 27 01:59:12 2017(r312855)
@@ -251,9 +251,9 @@ __DEFAULT_YES_OPTIONS+=LLVM_LIBUNWIND
__DEFAULT_NO_OPTIONS+=LLVM_LIBUNWIND
.endif
.if ${__T} == "aarch64"
-__DEFAULT_YES_OPTIONS+=LLD_AS_LD
+__DEFAULT_YES_OPTIONS+=LLD_IS_LD
.else
-__DEFAULT_NO_OPTIONS+=LLD_AS_LD
+__DEFAULT_NO_OPTIONS+=LLD_IS_LD
.endif
.if ${__T} == "aarch64" || ${__T} == "amd64"
__DEFAULT_YES_OPTIONS+=LLD LLDB
Copied: head/tools/build/options/WITHOUT_LLD_IS_LD (from r312854,
head/tools/build/options/WITHOUT_LLD_AS_LD)
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/tools/build/options/WITHOUT_LLD_IS_LD Fri Jan 27 01:59:12 2017
(r312855, copy of r312854, head/tools/build/options/WITHOUT_LLD_AS_LD)
@@ -0,0 +1,2 @@
+.\" $FreeBSD$
+Set to use GNU binutils ld as the system linker, instead of LLVM's LLD.
Copied: head/tools/build/options/WITH_LLD_IS_LD (from r312854,
head/tools/build/options/WITH_LLD_AS_LD)
==
--- /dev/null 00:00:00 1970 (empty, because file is newly added)
+++ head/tools/build/options/WITH_LLD_IS_LD Fri Jan 27 01:59:12 2017
(r312855, copy of r312854, head/tools/build/options/WITH_LLD_AS_LD)
@@ -0,0 +1,2 @@
+.\" $FreeBSD$
+Set to use LLVM's LLD as the system linker, instead of GNU binutils ld.
Modified: head/usr.bin/clang/lld/Makefile
==
--- head/usr.bin/clang/lld/Makefile Fri Jan 27 01:24:24 2017
(r312854)
+++ head/usr.bin/clang/lld/Makefile Fri Jan 27 01:59:12 2017
(r312855)
@@ -8,7 +8,7 @@ LLD_SRCS= ${LLVM_SRCS}/tools/lld
PACKAGE= lld
PROG_CXX= ld.lld
MAN=
-.if ${MK_LLD_AS_LD} != "no"
+.if ${MK_LLD_IS_LD} != "no"
SYMLINKS= ${PROG_CXX} ${BINDIR}/ld
.endif
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312857 - head
Author: emaste
Date: Fri Jan 27 03:43:18 2017
New Revision: 312857
URL: https://svnweb.freebsd.org/changeset/base/312857
Log:
Use cross-NM (XNM) in compat32 build
An attempt to build mips64 using external toolchain failed as it tried
to use the host amd64 nm.
MFC after:1 month
Sponsored by: The FreeBSD Foundation
Modified:
head/Makefile.libcompat
Modified: head/Makefile.libcompat
==
--- head/Makefile.libcompat Fri Jan 27 02:35:05 2017(r312856)
+++ head/Makefile.libcompat Fri Jan 27 03:43:18 2017(r312857)
@@ -50,6 +50,8 @@ LIB32WMAKEFLAGS= LD="${XLD} -m elf32btsm
LIB32WMAKEFLAGS+= OBJCOPY="${XOBJCOPY}"
.endif
+LIB32WMAKEFLAGS+= NM="${XNM}"
+
LIB32CFLAGS= -DCOMPAT_32BIT
LIB32DTRACE= ${DTRACE} -32
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
svn commit: r312859 - head/sys/arm/ti/am335x
Author: ian Date: Fri Jan 27 04:08:24 2017 New Revision: 312859 URL: https://svnweb.freebsd.org/changeset/base/312859 Log: Configure the timer capture pin to input mode in the timer control register, in addition to configuring it as input with the pinmux driver. There was a control register bit commented as "no desc in datasheet". A later revision of the manual reveals the bit to be an input/output control for the timer pin. In addition to configuring capture or pulse mode, you apparently have to separately configure the pin direction in the timer control register. Before this change, the timer block was apparently driving a signal onto a pad configured by pinmux as input. Capture mode still accidentally worked for me during testing because I was using a very strong signal source that just out-muscled the weaker drive from the misconfigured pin. Modified: head/sys/arm/ti/am335x/am335x_dmtpps.c head/sys/arm/ti/am335x/am335x_dmtreg.h Modified: head/sys/arm/ti/am335x/am335x_dmtpps.c == --- head/sys/arm/ti/am335x/am335x_dmtpps.c Fri Jan 27 03:44:50 2017 (r312858) +++ head/sys/arm/ti/am335x/am335x_dmtpps.c Fri Jan 27 04:08:24 2017 (r312859) @@ -463,6 +463,14 @@ dmtpps_attach(device_t dev) sc->tmr_num = ti_hwmods_get_unit(dev, "timer"); snprintf(sc->tmr_name, sizeof(sc->tmr_name), "DMTimer%d", sc->tmr_num); + /* +* Configure the timer pulse/capture pin to input/capture mode. This is +* required in addition to configuring the pin as input with the pinmux +* controller (which was done via fdt data or tunable at probe time). +*/ + sc->tclr = DMT_TCLR_GPO_CFG; + DMTIMER_WRITE4(sc, DMT_TCLR, sc->tclr); + /* Set up timecounter hardware, start it. */ DMTIMER_WRITE4(sc, DMT_TSICR, DMT_TSICR_RESET); while (DMTIMER_READ4(sc, DMT_TIOCP_CFG) & DMT_TIOCP_RESET) Modified: head/sys/arm/ti/am335x/am335x_dmtreg.h == --- head/sys/arm/ti/am335x/am335x_dmtreg.h Fri Jan 27 03:44:50 2017 (r312858) +++ head/sys/arm/ti/am335x/am335x_dmtreg.h Fri Jan 27 04:08:24 2017 (r312859) @@ -62,7 +62,7 @@ #define DMT_TCLR_TRGMODE_BOTH (2 << 10) /* Trigger on match + ovflow */ #define DMT_TCLR_PWM_PTOGGLE(1 << 12) /* PWM toggles */ #define DMT_TCLR_CAP_MODE_2ND (1 << 13) /* Capture second event mode */ -#define DMT_TCLR_GPO_CFG(1 << 14) /* (no descr in datasheet) */ +#define DMT_TCLR_GPO_CFG(1 << 14) /* Tmr pin conf, 0=out, 1=in */ #defineDMT_TCRR0x3C/* Counter Register */ #defineDMT_TLDR0x40/* Load Reg */ #defineDMT_TTGR0x44/* Trigger Reg */ ___ svn-src-head@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
Re: svn commit: r312770 - in head/sys: net netinet netinet6
On Thu, 26 Jan 2017, Gleb Smirnoff wrote:
On Thu, Jan 26, 2017 at 02:03:05PM +1100, Bruce Evans wrote:
B> On Thu, 26 Jan 2017, Konstantin Belousov wrote:
B>
B> > On Wed, Jan 25, 2017 at 02:20:06PM -0800, Gleb Smirnoff wrote:
B> >> Thanks, Luiz!
B> >>
B> >> One stylistic nit that I missed in review:
B> >>
B> >> L> static int
B> >> L> -in_difaddr_ioctl(caddr_t data, struct ifnet *ifp, struct thread *td)
B> >> L> +in_difaddr_ioctl(u_long cmd, caddr_t data, struct ifnet *ifp, struct
thread *td)
B> >> L> {
B> >> L>const struct ifreq *ifr = (struct ifreq *)data;
B> >> L>const struct sockaddr_in *addr = (const struct sockaddr_in *)
B> >> L> @@ -618,7 +618,8 @@ in_difaddr_ioctl(caddr_t data, struct if
B> >> L>in_ifadown(&ia->ia_ifa, 1);
B> >> L>
B> >> L>if (ia->ia_ifa.ifa_carp)
B> >> L> - (*carp_detach_p)(&ia->ia_ifa);
B> >> L> + (*carp_detach_p)(&ia->ia_ifa,
B> >> L> + (cmd == SIOCDIFADDR) ? false : true);
B> >>
B> >> Can we change the very last line to:
B> >>
B> >> (cmd == SIOCAIFADDR) ? true : false);
B>
B> That is not stylistic, but invert the result. Perhaps you meant to
B> reverse the test to avoid negative logic for the result.
It uses different ioctl value, so it doesn't invert result. Instead
of !SIOCDIFADDR I want more explicit SIOCAIFADDR.
Oops. So it is non-stylistic in a different way. cmd can only be
SIOCDIFADDR, or one or both of SIOCAIFADDR. Than is unclear. Assuming
that the original code is correct and that all 3 cases can occur,
inversion would break all 3 cases, while the non-stylistic change breaks
only the O_SIOCAIFADDR case.
Since there can be more than 2 cases and it isn't clear that there are
at most 3, any boolean test on 1 of the cases is going to be unclear.
Positive logic will be clearer, but that requires comparison with 2
cases. The current code use negative logic to select these 2 cases as
the complement of the other case.
Bruce
___
svn-src-head@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
